[{"data":1,"prerenderedAt":415},["ShallowReactive",2],{"explainers":3},{"data":4,"meta":410},[5,18,29,41,52,63,74,86,97,108,119,131,142,153,164,176,187,198,209,221,232,243,254,266,277,288,300,311,321,332,344,355,366,377,388,399],{"id":6,"attributes":7},99,{"title":8,"content":9,"createdAt":10,"updatedAt":11,"publishedAt":12,"date":13,"description":14,"keywords":15,"slug":16,"category":17},"Tools for protein structure analysis","Nanome is a collaborative molecular visualization and drug discovery platform that runs across a browser web app, XR headsets, and Windows desktop. For protein structure analysis, it loads PDB and SDF files (and pulls straight from RCSB PDB, PubChem, and DrugBank), and lets you inspect structures in real-time 3D or immersive XR. Its AI copilot, [MARA](https:\u002F\u002Fnanome.ai\u002Fmara), runs analyses like electrostatics, interactions, SASA, alignment, and pocket detection from a request written in plain English. It also plays well with the desktop tools most structural biologists already know.\n\n## What protein structure analysis actually involves\n\n![A researcher wearing an ultra-thin VR headset studies a ribbon-cartoon protein structure held close to the chest](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Ftools_for_protein_structure_analysis_image_1_v4s_748b4d5566.png)\n\n\n\n\"Structure analysis\" is a bundle of separate jobs that happen to share a file. Most projects touch several of them:\n\n- Loading a structure, usually PDB or mmCIF, often fetched by ID from RCSB PDB.\n- Reading secondary structure: where the helices and sheets sit, and how the loops between them pack.\n- Finding non-covalent contacts, hydrogen bonds and salt bridges among them, between residues or between a protein and a ligand.\n- Superimposing structures to compare conformations or homologs.\n- Detecting pockets and candidate binding sites.\n- Stepping through a molecular dynamics trajectory to watch the structure move.\n\nWhich software suits a group comes down to which of those jobs fills the week, and to how many people need to read the result together.\n\n## The common desktop tools\n\n\u003Cdiv class=\"logo-row\" style=\"display:flex;flex-wrap:wrap;align-items:center;justify-content:center;gap:1.75rem;background:#fff;border-radius:12px;padding:1.25rem 1.5rem;margin:2rem 0\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Ftools_for_protein_structure_analysis_image_6_6a381a491e.png\" alt='VMD' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Ftools_for_protein_structure_analysis_image_7_26bcbfe73e.png\" alt='Coot' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Ftools_for_protein_structure_analysis_image_8_dd32e0f13e.png\" alt='Schrödinger' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003C\u002Fdiv>\n\n\n\nFour names show up in nearly every structural biology group. All four are mature and scriptable, and free for academic work apart from PyMOL, whose maintained builds come through a paid academic subscription from Schrödinger.\n\n**[PyMOL](https:\u002F\u002Fpymol.org)** is the workhorse for publication figures and a quick look at a structure, with a command language and a Python API behind it for repeatable rendering. Nanome opens PyMOL `.pse` session files directly, so a saved session can be picked back up in immersive 3D. (QM\u002FMM link atoms can trip the parser, and the validated PyMOL session versions aren't documented.) That road runs one way: `.pse` is import only, and molecule export out of Nanome is PDB, SDF, or SMILES, single frame. [More on supported formats](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n\n**[UCSF ChimeraX](https:\u002F\u002Fwww.rbvi.ucsf.edu\u002Fchimerax\u002F)** (successor to Chimera) covers large assemblies, density maps, and deep analysis, with a wide command set and good performance on big structures. Nanome parses the same file set: PDB, mmCIF, SDF, MOL and MOL2, XYZ, and PQR, so a model prepared in ChimeraX opens with no conversion step in between.\n\n**[VMD](https:\u002F\u002Fwww.ks.uiuc.edu\u002FResearch\u002Fvmd\u002F)** grew up around molecular dynamics, and frame-by-frame motion analysis across a long trajectory is what it was built to do.\n\nFor simulation output, Nanome reads `.gro` on its own and attaches `.xtc`, `.trr`, and `.dcd` frames to a loaded model. Playback runs to the 2000 frame limit per trajectory. Surfaces are disabled while frames advance. [More on supported formats](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n\n**[Coot](https:\u002F\u002Fwww2.mrc-lmb.cam.ac.uk\u002Fpersonal\u002Fpemsley\u002Fcoot\u002F)** is where crystallographers fit a model into electron density and refine it. Nanome imports the PDB and mmCIF that come out the far end for inspection downstream. The density maps themselves stay in Coot, since CCP4, MRC, and DSN6 sit outside what Nanome parses. [More on supported formats](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n\nThese are desktop software, mostly one scientist at a keyboard, driven by a GUI or a script. That shape suits a great deal of the work, and Nanome doesn't set out to displace any of them.\n\n## Where Nanome fits\n\n![Two colleagues wearing ultra-thin VR headsets examine the same solid protein surface model with a visible binding pocket floating between them](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Ftools_for_protein_structure_analysis_image_2_v4_0d7755ced7.png)\n\n\n\nNanome adds two things on top of that stack: immersive inspection of the structure itself, and an AI copilot that drives the analysis tools.\n\nA PDB structure opens in real-time 3D in the browser, or you can step into it in XR on a [Quest, a Vive Focus 3, a Pico Neo, or an Apple Vision Pro](https:\u002F\u002Fnanome.ai\u002Fsetup). Several people can stand in the same structure at once and point at the same residue, which is hard to arrange over a screen share. Depth carries weight here too: a pocket that reads as a shallow dish on a flat display can turn out to have a lip and a back wall.\n\nThen there's MARA. You describe the analysis in ordinary words, and MARA picks what to run from a library of [300+ integrated scientific tools across 26 categories](https:\u002F\u002Fnanome.ai\u002Fintegrations). For protein structure work that covers:\n\n- Electrostatics through APBS.\n- [Non-covalent contacts between residues and ligands](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_v2\u002Ftoolspanel).\n- Solvent accessible surface area per residue.\n- Alignment and superposition.\n- Pocket and binding site detection.\n- Analysis of a molecular dynamics run.\n\nEach run comes back with its provenance attached: the tool that fired, the parameters it got, the output it returned. That gives a reviewer something to check rather than an answer to take on faith. For antibodies the method set narrows considerably, and [software for computational antibody design](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fsoftware-for-computational-antibody-design) goes through the numbering, CDR, and sequence design tools those projects lean on.\n\n## Comparison\n\n\u003Ctable class=\"table\">\n  \u003Cthead>\n    \u003Ctr>\u003Cth>Desktop tool\u003C\u002Fth>\u003Cth>Its strongest ground\u003C\u002Fth>\u003Cth>What Nanome puts next to it\u003C\u002Fth>\u003C\u002Ftr>\n  \u003C\u002Fthead>\n  \u003Ctbody>\n    \u003Ctr>\u003Ctd>PyMOL\u003C\u002Ftd>\u003Ctd>Publication figures, scripted rendering, a fast look at a structure\u003C\u002Ftd>\u003Ctd>Opens the saved .pse, then adds a room several people share and MARA running electrostatics, SASA, and pocket detection on request\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>UCSF ChimeraX\u003C\u002Ftd>\u003Ctd>Large assemblies, density maps, a deep command set\u003C\u002Ftd>\u003Ctd>Immersive 3D, and an analysis a scientist can ask for in words rather than type as a command\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>VMD\u003C\u002Ftd>\u003Ctd>Trajectory work and frame-by-frame motion analysis\u003C\u002Ftd>\u003Ctd>Playback a whole group watches together, with MARA reporting on the run\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Coot\u003C\u002Ftd>\u003Ctd>Model building and refinement into electron density\u003C\u002Ftd>\u003Ctd>Picks up the refined PDB or mmCIF for inspection and the analysis that follows\u003C\u002Ftd>\u003C\u002Ftr>\n  \u003C\u002Ftbody>\n\u003C\u002Ftable>\n\n## Where another tool is the better call\n\nIf the week is publication figures and refinement into density, PyMOL, ChimeraX, and Coot were built for exactly that and are worth keeping. Nanome suits the jobs where immersive 3D helps read a structure, where several people need to reach the same conclusion at the same time, or where asking for an analysis in words fits the group better than writing it. Plenty of labs run both. Nanome connects to [Schrödinger LiveDesign](https:\u002F\u002Fnanome.ai\u002Fintegrations), Cresset Flare, and [OpenEye](https:\u002F\u002Fwww.eyesopen.com) instead of asking anyone to drop them, and it reads [Maestro](https:\u002F\u002Fwww.schrodinger.com\u002Fplatform\u002Fproducts\u002Fmaestro\u002F) `.mae` and `.maegz`, the same files a LiveReport hands over.\n\nA wider survey of the viewers in this category sits in [the best molecular visualization tools](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fthe-best-molecular-visualization-tools). When the structure work feeds a chemistry program rather than ending at the figure, [drug discovery software for computational chemistry](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fdrug-discovery-software-for-computational-chemistry) maps the rest of that stack.\n\nAt UC San Diego, Prof. Zoran Radić's lab took a compound library aimed at nerve agent poisoning all the way from the X-ray structures at the start to lead optimization at the end, and published the work in the Journal of Biological Chemistry. Radić on what shifted: \"Virtual reality changed my perspective of macromolecules.\" More of that work is written up at [nanome.ai\u002Fcase-studies](https:\u002F\u002Fnanome.ai\u002Fcase-studies).\n\n## FAQ\n\n**What file formats does Nanome load for protein structure analysis?**\nStructures: PDB (`.pdb`, `.ent`), mmCIF (`.cif`, `.mmcif`, `.mcif`, `.bcif`), SDF, MOL and MOL2, XYZ, PQR, SMILES, and PDBQT (converted to PDB on load, with charges dropped). Vendor and session files: Maestro `.mae` and `.maegz`, MOE `.moe`, and PyMOL `.pse`, all import only. Trajectories: `.gro` loads on its own, while `.xtc`, `.trr`, and `.dcd` attach to a model already open and have to match its atom count. Electrostatic maps come in as `.dx` overlays on a loaded model. Structures also arrive by ID from RCSB PDB, PubChem, and DrugBank. What comes back out is PDB, SDF, or SMILES, single frame. [More on supported formats](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n\n**Can pockets be found without writing a script?**\nYes. A plain English request to MARA runs pocket and binding site detection through the integrated tools, and the reply names the tool that ran and what it returned.\n\n**Does Nanome replace PyMOL or ChimeraX?**\nNo. It sits beside them. What it adds is a session several people join at once, native XR inspection, and MARA taking analysis requests in ordinary words. It also connects to several established suites rather than competing across the board.\n\n**Is a headset required?**\nNo. The browser web app runs without one. XR is there on Apple Vision Pro, on Quest, on Vive Focus 3, and on Pico Neo, for the days when immersive 3D earns the extra step.\n","2026-07-15T01:23:51.976Z","2026-09-24T16:00:05.194Z","2026-09-24T16:00:05.141Z","2026-09-24","The best tools for protein structure analysis, from PyMOL and ChimeraX to Nanome's interactive 3D and XR inspection with MARA agents.","protein structure analysis tools, best tools for protein structure analysis, PDB viewer, protein visualization, binding site detection, molecular dynamics analysis, Nanome, MARA","tools-for-protein-structure-analysis","frequent-topics",{"id":19,"attributes":20},104,{"title":21,"content":22,"createdAt":23,"updatedAt":24,"publishedAt":25,"date":13,"description":26,"keywords":27,"slug":28,"category":17},"VR tools for molecular modeling","Nanome is a collaborative molecular visualization and drug discovery platform built for VR and XR. You load a protein or small molecule, then walk around it, grab it, and reshape it at true 3D scale with your hands. It runs on [major headsets, and there's a browser web app](https:\u002F\u002Fnanome.ai\u002Fsetup) for anyone without one, so a whole team can join the same structure at once.\n\nA monitor draws a 3D object as a flat projection, and the depth has to be rebuilt by spinning the view. In a headset the depth arrives with the molecule.\n\n## What VR adds to molecular modeling\n\n![A researcher wearing an ultra-thin VR headset studies a ribbon-cartoon protein structure held close to the chest](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fvr_tools_for_molecular_modeling_image_1_v4_b5fc23159c.png)\n\n\n\nFour things change once the structure is an object in the room rather than a picture of one.\n\n**Depth.** Stereo vision puts every atom at its real distance. A binding pocket reads as a cavity with a floor and walls, and a subpocket behind the ligand stops hiding.\n\n**Scale.** A ligand can sit in your palm while you check a torsion angle. A minute later the same protein can fill the room and you can stand at the mouth of the active site.\n\n**Hands.** Rotating, translating and adjusting a structure happens by moving your hands through the space it occupies, with no keyboard modifier between the gesture and the motion.\n\n**Company.** Several people hold one workspace at the same time. A chemist in San Diego and a biologist in Boston can point at the same residue, and the pointing lands on the atom itself instead of on a description of it.\n\n\u003Ctable class=\"table\">\n  \u003Cthead>\n    \u003Ctr>\u003Cth>What changes\u003C\u002Fth>\u003Cth>On a flat screen\u003C\u002Fth>\u003Cth>In Nanome VR\u002FXR\u003C\u002Fth>\u003C\u002Ftr>\n  \u003C\u002Fthead>\n  \u003Ctbody>\n    \u003Ctr>\u003Ctd>Depth perception\u003C\u002Ftd>\u003Ctd>Rebuilt by rotating the view\u003C\u002Ftd>\u003Ctd>Seen directly in stereo 3D\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Scale\u003C\u002Ftd>\u003Ctd>A zoom level\u003C\u002Ftd>\u003Ctd>Anything from palm-sized to room-sized\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Manipulation\u003C\u002Ftd>\u003Ctd>Mouse plus keyboard modifiers\u003C\u002Ftd>\u003Ctd>Both hands on the structure\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Collaboration\u003C\u002Ftd>\u003Ctd>Screen-share with one driver\u003C\u002Ftd>\u003Ctd>Everyone in the same room around one molecule\u003C\u002Ftd>\u003C\u002Ftr>\n  \u003C\u002Ftbody>\n\u003C\u002Ftable>\n\n## Which devices run it\n\n\u003Ctable class=\"table\">\n  \u003Cthead>\n    \u003Ctr>\u003Cth>Device\u003C\u002Fth>\u003Cth>Type\u003C\u002Fth>\u003Cth>Headset needed\u003C\u002Fth>\u003C\u002Ftr>\n  \u003C\u002Fthead>\n  \u003Ctbody>\n    \u003Ctr>\u003Ctd>Meta Quest\u003C\u002Ftd>\u003Ctd>VR headset\u003C\u002Ftd>\u003Ctd>Yes\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>HTC Vive Focus 3\u003C\u002Ftd>\u003Ctd>VR headset\u003C\u002Ftd>\u003Ctd>Yes\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Pico Neo\u003C\u002Ftd>\u003Ctd>VR headset\u003C\u002Ftd>\u003Ctd>Yes\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Apple Vision Pro\u003C\u002Ftd>\u003Ctd>XR headset\u003C\u002Ftd>\u003Ctd>Yes\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Windows desktop\u003C\u002Ftd>\u003Ctd>Desktop app\u003C\u002Ftd>\u003Ctd>No\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Web app (browser)\u003C\u002Ftd>\u003Ctd>Browser\u003C\u002Ftd>\u003Ctd>No\u003C\u002Ftd>\u003C\u002Ftr>\n  \u003C\u002Ftbody>\n\u003C\u002Ftable>\n\nThat last row does the most work. Headsets are rarely one per desk, and a review can go ahead regardless: a medicinal chemist joins from a laptop while a colleague works in full immersion, and the two of them are in one live session on one structure. Running a whole group that way, across sites and clocks, gets a fuller treatment in [collaborative drug discovery software for remote teams](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fcollaborative-drug-discovery-software-for-remote-teams).\n\nStructures arrive by accession code from RCSB PDB, PubChem, DrugBank, UniProt, ChEMBL and AlphaFold DB, or straight off a drive, so the molecule is in the room a few seconds after the session opens.\n\n## Running real tools inside the headset\n\n![A plain-English voice command flows through a computational tool and returns a docked ligand result in one connected step.](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fvr_tools_for_molecular_modeling_image_2_265c9cc549.png)\n\n\n\nSeeing a molecule in 3D is half the job. The other half is the calculation that says whether an idea survives contact with a number, and [MARA](https:\u002F\u002Fnanome.ai\u002Fmara), Nanome's AI copilot, covers that half without anyone leaving the structure.\n\nYou describe the job in plain English. MARA chooses the tool, runs it, and keeps a record of the run: the tool by name, the structure it was handed, the file that came back. [300+ tools across 26 categories](https:\u002F\u002Fnanome.ai\u002Fintegrations) sit behind that one request, and a REST API plus MCP servers open the same catalogue to scripts.\n\nSome of what it can call:\n\n- Docking on Smina and DiffDock-L\n- Electrostatics through APBS\n- ADMET and toxicity prediction\n- Fold and complex prediction with [AlphaFold 3](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41586-024-07487-w), [Boltz-2](https:\u002F\u002Fgithub.com\u002Fjwohlwend\u002Fboltz), OpenFold3 and Chai-1\n- Sequence design with ProteinMPNN, plus ANARCI for antibody numbering and CDR loops\n- De novo binder design with RFdiffusion3 (beta)\n- Cheminformatics\n\nIn a headset the request can be spoken out loud. Both hands stay on the structure, you say what you want run, and the result comes back onto the molecule you're still holding.\n\nThat makes docking something you do where the structure already is. Ask for a compound to be placed in the pocket you're standing inside, then turn the pose over and read the contacts yourself. What a careful pass over those contacts involves is the subject of [how to analyze protein-ligand interactions](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fhow-to-analyze-protein-ligand-interactions).\n\n## How this sits next to the desktop tools\n\n\u003Cdiv class=\"logo-row\" style=\"display:flex;flex-wrap:wrap;align-items:center;justify-content:center;gap:1.75rem;background:#fff;border-radius:12px;padding:1.25rem 1.5rem;margin:2rem 0\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fvr_tools_for_molecular_modeling_image_4_3fdd81cf89.png\" alt='PyMOL' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fvr_tools_for_molecular_modeling_image_6_7179e07eb8.png\" alt='VMD' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fvr_tools_for_molecular_modeling_image_7_70bed70d36.png\" alt='Schrödinger Maestro' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fvr_tools_for_molecular_modeling_image_8_0a03b792fd.png\" alt='MOE' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003C\u002Fdiv>\n\n\n\n![Two colleagues in casual professional attire discuss a space-filling protein structure on a large wall display in a modern research lounge.](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fvr_tools_for_molecular_modeling_image_3_fd3296840c.png)\n\n\n\nA lot of excellent modeling software was written for one scientist at one keyboard, and it is very good at that. Here is how the pieces divide up.\n\n\u003Ctable class=\"table\">\n  \u003Cthead>\n    \u003Ctr>\u003Cth>Tool\u003C\u002Fth>\u003Cth>Its strength\u003C\u002Fth>\u003Cth>What immersive 3D adds beside it\u003C\u002Fth>\u003C\u002Ftr>\n  \u003C\u002Fthead>\n  \u003Ctbody>\n    \u003Ctr>\u003Ctd>\u003Ca href=\"https:\u002F\u002Fpymol.org\">PyMOL\u003C\u002Fa>, \u003Ca href=\"https:\u002F\u002Fwww.rbvi.ucsf.edu\u002Fchimerax\u002F\">ChimeraX\u003C\u002Fa>, \u003Ca href=\"https:\u002F\u002Fwww.ks.uiuc.edu\u002FResearch\u002Fvmd\u002F\">VMD\u003C\u002Fa>\u003C\u002Ftd>\u003Ctd>Desktop rendering, scripting, figures for a paper\u003C\u002Ftd>\u003Ctd>A structure a group can walk around together, live\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>\u003Ca href=\"https:\u002F\u002Favogadro.cc\">Avogadro\u003C\u002Fa>, \u003Ca href=\"https:\u002F\u002Fwww2.mrc-lmb.cam.ac.uk\u002Fpersonal\u002Fpemsley\u002Fcoot\u002F\">Coot\u003C\u002Fa>\u003C\u002Ftd>\u003Ctd>Building and correcting geometry\u003C\u002Ftd>\u003Ctd>Editing by hand at the scale of the pocket itself\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>\u003Ca href=\"https:\u002F\u002Fwww.schrodinger.com\u002Fplatform\u002Fproducts\u002Fmaestro\u002F\">Schrödinger Maestro\u003C\u002Fa>, \u003Ca href=\"https:\u002F\u002Fwww.chemcomp.com\">MOE\u003C\u002Fa>, \u003Ca href=\"https:\u002F\u002Fwww.3ds.com\u002Fproducts\u002Fbiovia\u002Fdiscovery-studio\">BIOVIA Discovery Studio\u003C\u002Fa>\u003C\u002Ftd>\u003Ctd>Full desktop comp-chem suites\u003C\u002Ftd>\u003Ctd>Several are Nanome integrations, and their output opens in a shared room\u003C\u002Ftd>\u003C\u002Ftr>\n  \u003C\u002Ftbody>\n\u003C\u002Ftable>\n\nTheir files cross over without a conversion step in between. Nanome opens PyMOL `.pse` sessions, Maestro `.mae` and `.maegz`, and MOE `.moe` files, all three as imports. The everyday structure files open too, which is what ChimeraX, VMD, Coot and Discovery Studio write: PDB, mmCIF, SDF, MOL and MOL2, XYZ and PQR. [More on supported formats](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n\nPipelines connect as well. [Schrödinger LiveDesign](https:\u002F\u002Fnanome.ai\u002Fintegrations), Cresset Flare, [CDD Vault](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fcollaborative-drug-discovery-and-nanome-partnership-announcement), [OpenEye \u002F Cadence](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fview-your-openeyecadence-data-with-fresh-eyes-using-nanome-xr), the [OpenFold Consortium](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fnanome-joins-the-openfold-consortium), [KNIME](https:\u002F\u002Fwww.knime.com) and Jupyter all feed it. When the numbers get produced in one of those, Nanome is where the group stands around the result.\n\nSome work belongs elsewhere. A headless screen across a million compounds belongs in a script, and a single static figure for a journal belongs in a desktop renderer. Immersion pays when one person has to understand a 3D arrangement, or when several people have to agree about one. Carrying that agreement to colleagues who were never in the session is a separate craft, worked through in [what a modern molecular presentation looks like](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fwhat-a-modern-molecular-presentation-looks-like).\n\nChemistry designed this way does leave the headset. Oak Ridge National Laboratory built a new inhibitor of the SARS-CoV-2 main protease inside Nanome, adding a chlorine atom that bound the protease more tightly, and the compound showed superior inhibition in vitro. The Journal of Medicinal Chemistry carries the work, and if it clears further development it will be the first ever drug discovered in virtual reality. That paper and the rest of the peer-reviewed record sit under [publications](https:\u002F\u002Fnanome.ai\u002Fpublications), and the project write-ups are at [nanome.ai\u002Fcase-studies](https:\u002F\u002Fnanome.ai\u002Fcase-studies).\n\n## FAQ\n\n**What is the best VR tool for molecular modeling?**\nNanome is purpose-built for it. It runs on Apple Vision Pro, Meta Quest, Pico Neo and HTC Vive Focus 3, plus a Windows desktop app and a browser web app for anyone off-headset, and several people share one live structure at a time. MARA runs the computational tools inside that same session.\n\n**Can I use Nanome without a headset?**\nYes. The browser web app and the Windows desktop app both join the same session, so a colleague on a laptop takes part fully while a teammate works in full immersion.\n\n**Can I run computational chemistry in VR, or only look at molecules?**\nReal tools run. MARA takes the request in plain English and calls docking, co-folding, electrostatics, ADMET or structure prediction, then names the tool it used and what came back. 300+ tools across 26 categories are reachable that way.\n\n**What file formats and databases does Nanome support?**\nStructures import as PDB (`.pdb`, `.ent`), mmCIF (`.cif`, `.mmcif`), SDF, MOL and MOL2, SMILES, XYZ, PQR, and PDBQT (converted to PDB, with charges dropped). Vendor and session files import as well: PyMOL `.pse`, Maestro `.mae` and `.maegz`, MOE `.moe`. An electrostatic map (`.dx`) overlays onto a structure that is already open. Export is PDB, SDF or SMILES, single frame, which leaves mmCIF, MAE, MOE and PSE import-only. Databases: fetch by code from RCSB PDB, PubChem, DrugBank, UniProt, ChEMBL and AlphaFold DB. Full detail is in [the format docs](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n","2026-07-15T01:23:52.433Z","2026-09-24T16:00:06.910Z","2026-09-24T16:00:06.850Z","VR tools for molecular modeling let you build and study molecules in 3D. Nanome is the purpose-built VR\u002FXR platform, with a web app too.","VR tools for molecular modeling, VR molecular modeling, XR drug discovery, molecular visualization VR, Nanome, immersive molecular modeling","vr-tools-for-molecular-modeling",{"id":30,"attributes":31},81,{"title":32,"content":33,"createdAt":34,"updatedAt":35,"publishedAt":36,"date":37,"description":38,"keywords":39,"slug":40,"category":17},"The best molecular modeling software","The best molecular modeling software depends on what you're modeling and how you want to touch it. Nanome is a collaborative platform where you build, edit, and minimize molecular structures in 3D across [a web app and XR headsets](https:\u002F\u002Fnanome.ai\u002Fsetup). Inside it, [an AI copilot called MARA](https:\u002F\u002Fnanome.ai\u002Fmara) handles the simulation-adjacent work: minimization, conformer generation, and docking. For classic desktop building and editing there's [Avogadro](https:\u002F\u002Favogadro.cc), and for full comp-chem suites there's [MOE](https:\u002F\u002Fwww.chemcomp.com), [Schrödinger](https:\u002F\u002Fwww.schrodinger.com), and [BIOVIA Discovery Studio](https:\u002F\u002Fwww.3ds.com\u002Fproducts\u002Fbiovia\u002Fdiscovery-studio).\n\nMolecular modeling is a wide category. It covers building a structure atom by atom, editing bonds and geometry, minimizing energy to reach a sensible conformation, and running the physics that predicts how a molecule behaves. A given project usually needs one or two of those and rarely all four, so the right pick tracks the specific job.\n\n## What to look for\n\n![A researcher wearing an ultra-thin VR headset studies a ribbon-cartoon protein structure held close to the chest](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fthe_best_molecular_modeling_software_image_1_v4_3618f58feb.png)\n\n\n\nThe task comes first. Sketching a small molecule, cleaning up a protein, generating conformers, and setting up a docking run pull in different directions, and few packages are the strongest option for all four.\n\nInputs come next. Modeling software that reads PDB for proteins and SDF for small molecules, and fetches structures straight out of RCSB PDB and PubChem, takes a conversion step off the front of the job.\n\nThen there's how the structure gets handled once it's open. A mouse and a flat panel are fine for 2D sketching. Fitting a ligand into a binding pocket is a judgment about depth and clearance, and depth is the first thing a flat projection throws away.\n\n## Where each tool fits\n\n\u003Ctable class=\"table\">\n  \u003Cthead>\n    \u003Ctr>\u003Cth>Software\u003C\u002Fth>\u003Cth>Its strength\u003C\u002Fth>\u003Cth>How Nanome pairs with it\u003C\u002Fth>\u003C\u002Ftr>\n  \u003C\u002Fthead>\n  \u003Ctbody>\n    \u003Ctr>\n      \u003Ctd>Avogadro\u003C\u002Ftd>\n      \u003Ctd>Free desktop builder and editor for small molecules, with force-field minimization\u003C\u002Ftd>\n      \u003Ctd>Nanome runs the same build-and-minimize loop at arm's length in 3D, with several people in the workspace at once. It imports the ordinary structure files: PDB, mmCIF, SDF, MOL2, XYZ and PQR, plus PDBQT since 2.6.0.\u003C\u002Ftd>\n    \u003C\u002Ftr>\n    \u003Ctr>\n      \u003Ctd>MOE (Molecular Operating Environment)\u003C\u002Ftd>\n      \u003Ctd>Full comp-chem suite: modeling, docking, protein prep, cheminformatics\u003C\u002Ftd>\n      \u003Ctd>CCG MOE is a listed \u003Ca href=\"https:\u002F\u002Fnanome.ai\u002Fintegrations\">Nanome integration\u003C\u002Fa>, so the two sit in one chain. Nanome opens \u003Ccode>.moe\u003C\u002Fcode> files for viewing, and there's no \u003Ccode>.moe\u003C\u002Fcode> writer going back.\u003C\u002Ftd>\n    \u003C\u002Ftr>\n    \u003Ctr>\n      \u003Ctd>Schrödinger (\u003Ca href=\"https:\u002F\u002Fwww.schrodinger.com\u002Fplatform\u002Fproducts\u002Fmaestro\u002F\">Maestro\u003C\u002Fa>, LiveDesign)\u003C\u002Ftd>\n      \u003Ctd>Physics-based modeling and simulation at scale\u003C\u002Ftd>\n      \u003Ctd>Nanome connects to \u003Ca href=\"https:\u002F\u002Fnanome.ai\u002Fintegrations\">LiveDesign\u003C\u002Fa> and opens Maestro \u003Ccode>.mae\u003C\u002Fcode> and \u003Ccode>.maegz\u003C\u002Fcode>, the format LiveDesign hands structures off in. The heavy compute stays in Schrödinger.\u003C\u002Ftd>\n    \u003C\u002Ftr>\n    \u003Ctr>\n      \u003Ctd>BIOVIA Discovery Studio\u003C\u002Ftd>\n      \u003Ctd>Broad life-science modeling and simulation environment\u003C\u002Ftd>\n      \u003Ctd>Nanome sits beside the suite as the hands-on 3D layer for building and minimizing, and takes the same structure files the suite already writes.\u003C\u002Ftd>\n    \u003C\u002Ftr>\n  \u003C\u002Ftbody>\n\u003C\u002Ftable>\n\nTraffic the other way is narrower. Nanome writes PDB, SDF or SMILES, one frame at a time, which leaves mmCIF, MAE and MOE as import-only. [More on supported formats](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n\n## What Nanome does\n\n![Two colleagues review a ligand docked into a protein binding pocket on a large wall display in a modern lounge.](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fthe_best_molecular_modeling_software_image_2_bb98f48c3c.png)\n\n\n\nYou build and edit at close to hand scale. Grab an atom, change a bond, adjust a torsion, then run a minimization and let the geometry settle.\n\nHeadsets covered: Meta Quest, Apple Vision Pro, HTC Vive Focus 3, and Pico Neo. There's a Windows desktop build as well, and a browser app that asks for nothing beyond a laptop.\n\nStructures arrive from a local file or straight out of RCSB PDB, PubChem, and DrugBank. Several people can hold the same molecule in the same workspace, so an edit one person makes shows up for everyone else while it happens.\n\nMARA covers the compute that sits next to modeling. Describe the job in plain English and it minimizes, generates conformers, and sets up docking with Smina or DiffDock-L, alongside [co-folding, electrostatics via APBS, ADMET prediction, and structure prediction](https:\u002F\u002Fnanome.ai\u002Fagents) using [AlphaFold 3](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41586-024-07487-w), [Boltz-2](https:\u002F\u002Fgithub.com\u002Fjwohlwend\u002Fboltz), and OpenFold3. That library runs to [300+ tools across 26 categories](https:\u002F\u002Fnanome.ai\u002Fintegrations), and every result comes back carrying its tool name, its inputs, and its raw output, so a number can be traced. On the antibody side, ANARCI numbers and classifies variable domains and ProteinMPNN designs sequences, which is the whole subject of [software for computational antibody design](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fsoftware-for-computational-antibody-design).\n\nWhen the modeling ends in molecular dynamics, the trajectory comes along with it. For simulation output, Nanome reads `.gro` on its own and attaches `.xtc`, `.trr`, and `.dcd` frames to a loaded model. Playback runs to a 2000-frame cap per trajectory. Surfaces are disabled while frames advance. [More on supported formats](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n\nMolecules built in the headset get made. Researchers at Oak Ridge National Laboratory worked inside the structure with the MedChem plug-in and added a chlorine atom to a SARS-CoV-2 Mpro inhibitor, and the modified compound bound the protease better and showed stronger inhibition in vitro. The result was published in the Journal of Medicinal Chemistry. First author Dr. Daniel Kneller described it this way: \"This novel chemical structure is different from what has been previously studied by the global community.\"\n\n## When a different tool is the better fit\n\n![A flat diagram showing four tool categories in a row, with the collaborative spatial layer highlighted as the connecting step.](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fthe_best_molecular_modeling_software_image_3_404aaa4c0c.png)\n\n\n\nSketching one small molecule on a laptop and minimizing it is Avogadro's home ground, and it costs nothing. Scripted physics at high volume is what Schrödinger and MOE were built for, and that trade-off gets a longer treatment in [drug discovery software for computational chemistry](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fdrug-discovery-software-for-computational-chemistry). When the job is looking rather than building, the [roundup of molecular visualization tools](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fthe-best-molecular-visualization-tools) covers the viewers.\n\nNanome's half of the week is building, editing, and reasoning about structures in space, with colleagues in the room and the simulation-adjacent steps handed to MARA. It plugs into those suites instead of standing in for them. The [case studies](https:\u002F\u002Fnanome.ai\u002Fcase-studies) show what that looked like on live programs.\n\n## FAQ\n\n**What is molecular modeling software?**\nSoftware for building, editing, minimizing, and simulating molecular structures. Some packages specialize in a single step, and full suites carry the chain from structure through to physics.\n\n**What file formats does Nanome support?**\nImport covers the ordinary structure formats: PDB (`.pdb`, `.ent`), mmCIF (`.cif`, `.mmcif`, `.bcif`), SDF (`.sdf`, `.mol`), MOL\u002FMOL2, SMILES, XYZ, PQR, and PDBQT (added in 2.6.0, converted to PDB with charges dropped). Vendor and session files come in too: Maestro `.mae` and `.maegz`, MOE `.moe`, and PyMOL `.pse`. On trajectories, `.gro` loads by itself, while `.xtc`, `.trr`, and `.dcd` attach to a model that's already open and have to match its atom count. Electrostatic `.dx` maps overlay a loaded model. Writing back out is narrower: PDB, SDF or SMILES, single frame, which leaves mmCIF, MAE, MOE and PSE import-only. [More on supported formats](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n\n**Can I model molecules in VR?**\nYes. Nanome runs on Meta Quest, Apple Vision Pro, HTC Vive Focus 3, and Pico Neo, so building, editing, and minimizing happen at true 3D scale. The browser app opens the same structures for anyone without a headset.\n\n**Does Nanome need a headset?**\nNo. It runs in a browser and on Windows desktop, and one session can mix those with people who are in headsets.\n\n**Does Nanome replace Schrödinger or MOE?**\nIt works alongside them. MOE, Schrödinger LiveDesign, and CDD Vault are all listed integrations, so structures move between Nanome and the suite while the heavy compute stays where it already runs.\n","2026-07-15T01:23:50.009Z","2026-09-22T16:00:07.248Z","2026-09-22T16:00:07.141Z","2026-09-22","The best molecular modeling software depends on your task. Nanome lets you build, edit, and minimize structures in 3D and XR.","best molecular modeling software, molecular modeling, molecular modeling tools, build and edit molecules, molecular minimization, molecular modeling in VR, Nanome","the-best-molecular-modeling-software",{"id":42,"attributes":43},101,{"title":44,"content":45,"createdAt":46,"updatedAt":47,"publishedAt":48,"date":37,"description":49,"keywords":50,"slug":51,"category":17},"The best tools for structure-based drug design","Structure-based drug design (SBDD) means using a target protein's 3D structure to design molecules that bind it. The best tools depend on the step you're on: full comp-chem suites like [Schrödinger](https:\u002F\u002Fwww.schrodinger.com) [Maestro](https:\u002F\u002Fwww.schrodinger.com\u002Fplatform\u002Fproducts\u002Fmaestro\u002F), [MOE](https:\u002F\u002Fwww.chemcomp.com), and [OpenEye](https:\u002F\u002Fwww.eyesopen.com) handle docking and optimization, while **Nanome** adds a collaborative visualization and AI copilot layer ([MARA](https:\u002F\u002Fnanome.ai\u002Fmara)) that runs docking, de novo binders, and electrostatics right on top of them. Nanome loads PDB and SDF structures across a browser, XR headsets, and Windows desktop, and plugs into the suites you already run.\n\n## What is structure-based drug design?\n\n![A researcher wearing an ultra-thin VR headset studies a solid protein surface model with a small ligand visible in its binding pocket held close to the chest](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fthe_best_tools_for_structure_based_drug_design_image_1_v4s_11c0969067.png)\n\n\n\nSBDD starts with a 3D structure of the target, usually a protein, and works out what small molecule or biologic fits its binding site. You look at the pocket, place candidate ligands, score how well they bind, then refine.\n\nLigand-based design goes the other direction, inferring a shape from molecules already known to work. With the pocket in front of you, hydrogen bonds, shape, and charge become things you can measure and argue about directly.\n\n## The workflow, step by step\n\nMost SBDD projects move through 4 stages, and the package that suits one stage rarely suits all four.\n\n1. **Get the structure.** Pull an experimental structure from RCSB PDB, or predict one with [AlphaFold 3](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41586-024-07487-w) or [Boltz-2](https:\u002F\u002Fgithub.com\u002Fjwohlwend\u002Fboltz) when no crystal exists.\n2. **Find the pocket.** Identify the binding site and map its shape, hydrophobic patches, and charged residues.\n3. **Dock.** Place candidate molecules in the pocket and score the poses.\n4. **Optimize.** Iterate on the best hits: tweak substituents, check ADMET, design de novo binders.\n\n## Comparison table\n\n\u003Ctable class=\"table\">\n  \u003Cthead>\n    \u003Ctr>\u003Cth>Package\u003C\u002Fth>\u003Cth>Strength\u003C\u002Fth>\u003Cth>How Nanome pairs with it\u003C\u002Fth>\u003C\u002Ftr>\n  \u003C\u002Fthead>\n  \u003Ctbody>\n    \u003Ctr>\u003Ctd>Schrödinger Maestro \u002F LiveDesign\u003C\u002Ftd>\u003Ctd>Physics-based docking, free-energy perturbation, enterprise data\u003C\u002Ftd>\u003Ctd>Nanome imports Maestro \u003Ccode>.mae\u003C\u002Fcode> and \u003Ccode>.maegz\u003C\u002Fcode> files, the same format LiveDesign ingests, and connects to \u003Ca href=\"https:\u002F\u002Fnanome.ai\u002Fintegrations\">LiveDesign\u003C\u002Fa>, so a compound in the LiveReport comes up in shared 3D. \u003Ca href=\"https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats\">Formats in detail\u003C\u002Fa>\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>MOE (Molecular Operating Environment)\u003C\u002Ftd>\u003Ctd>Structure prep, pharmacophores, medicinal chemistry workflows\u003C\u002Ftd>\u003Ctd>MOE is a listed Nanome integration. Nanome reads \u003Ccode>.moe\u003C\u002Fcode> files, and MARA can dock or compute electrostatics on that same structure without a hand-off\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>OpenEye (Cadence)\u003C\u002Ftd>\u003Ctd>Shape and electrostatic similarity, fast virtual screening\u003C\u002Ftd>\u003Ctd>Nanome integrates with \u003Ca href=\"https:\u002F\u002Fnanome.ai\u002Fblog\u002Fview-your-openeyecadence-data-with-fresh-eyes-using-nanome-xr\">OpenEye\u003C\u002Fa>, then puts the hit list in front of several people at once for a live pass\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Cresset Flare\u003C\u002Ftd>\u003Ctd>Electrostatics-driven design, field-based analysis\u003C\u002Ftd>\u003Ctd>Nanome connects to Flare and puts the fields and the ligand in a room a group can walk around\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>PyMOL, ChimeraX, VMD\u003C\u002Ftd>\u003Ctd>Desktop, script-driven structure viewing\u003C\u002Ftd>\u003Ctd>Nanome opens a PyMOL \u003Ccode>.pse\u003C\u002Fcode> session for viewing (import only, and nothing writes back out as \u003Ccode>.pse\u003C\u002Fcode>), along with \u003Ccode>.pdb\u003C\u002Fcode>, \u003Ccode>.cif\u003C\u002Fcode>, \u003Ccode>.sdf\u003C\u002Fcode>, \u003Ccode>.mol2\u003C\u002Fcode>, \u003Ccode>.xyz\u003C\u002Fcode>, and \u003Ccode>.pqr\u003C\u002Fcode>. Saving a molecule back out means PDB, SDF, or SMILES, single-frame. \u003Ca href=\"https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats\">Formats in detail\u003C\u002Fa>\u003C\u002Ftd>\u003C\u002Ftr>\n  \u003C\u002Ftbody>\n\u003C\u002Ftable>\n\n## Where Nanome and MARA fit\n\n![Two colleagues wearing ultra-thin VR headsets examine the same ribbon-cartoon protein structure floating between them](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fthe_best_tools_for_structure_based_drug_design_image_2_v4s_40d53d5c1b.png)\n\n\n\nNanome is a collaborative molecular visualization and drug discovery platform. It opens in a browser, on Windows desktop, and in [XR headsets](https:\u002F\u002Fnanome.ai\u002Fsetup), and a structure looks the same in all three.\n\nThe AI copilot inside it is **MARA**. Describe a job in plain English, and MARA picks a tool, runs it, then names what it called, what went in, and what came back. Its built-in library covers 26 categories, and the [full catalog](https:\u002F\u002Fnanome.ai\u002Fintegrations) is public.\n\nEvery one of the 4 stages above has engines waiting in there. Structure prediction and co-folding go through AlphaFold 3, Boltz-2, OpenFold3, or Chai-1. Docking runs on Smina or DiffDock-L. [APBS handles electrostatics, ADMET models score the survivors, and RFdiffusion3 (beta) builds de novo binders](https:\u002F\u002Fnanome.ai\u002Fagents). ProteinMPNN writes sequences onto a backbone, while ANARCI numbers and classifies antibody variable domains and marks their CDR loops. Structures come in from RCSB PDB, PubChem, and DrugBank.\n\nAntibodies bend every one of those stages into a different shape, and [software for computational antibody design](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fsoftware-for-computational-antibody-design) follows that version of the pipeline.\n\nReview is where the shared room does the most. A medicinal chemist, a structural biologist, and a computational chemist can stand around one pocket at the same time, so a question about which residue somebody means gets settled by pointing at it.\n\nPast the four packages in the table, Nanome links to [Collaborative Drug Discovery (CDD Vault)](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fcollaborative-drug-discovery-and-nanome-partnership-announcement), [KNIME](https:\u002F\u002Fwww.knime.com), and Jupyter, and exposes a REST API plus MCP servers, so it drops into a pipeline rather than asking for a new one.\n\n## When to use something else\n\n![A researcher at a plain desk studies a space-filling protein model on a large monitor in a quiet computational office.](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fthe_best_tools_for_structure_based_drug_design_image_3_bf599faaa1.png)\n\n\n\nFree-energy calculations at production scale are a suite job, and Schrödinger is built for that arithmetic, with Nanome sitting above it for the looking and the deciding. A quick scripted edit on one structure is lighter work for a desktop viewer like [PyMOL](https:\u002F\u002Fpymol.org) or [ChimeraX](https:\u002F\u002Fwww.rbvi.ucsf.edu\u002Fchimerax\u002F). Nanome pays off when several people need the same pocket in front of them at once, or when an AI copilot is doing the running.\n\nViewers as a category get a head-to-head of their own in [the best molecular visualization tools](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fthe-best-molecular-visualization-tools), and the wider stack around these 4 stages, from cheminformatics to data management, is mapped in [drug discovery software for computational chemistry](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fdrug-discovery-software-for-computational-chemistry).\n\nSome of this shows up in the literature. A team at Oak Ridge National Laboratory worked inside Nanome with its MedChem plug-in and built a new inhibitor of the SARS-CoV-2 main protease, Mpro. Hanging a chlorine atom off the scaffold made it bind the protease better, and the compound showed superior inhibition in vitro. The paper ran in the Journal of Medicinal Chemistry, where first author Dr. Kneller described the chemical structure as different from what the global community had studied before. If it clears further development, it stands as a candidate for the first drug anyone has discovered inside virtual reality. [Nanome's case studies](https:\u002F\u002Fnanome.ai\u002Fcase-studies) carry the write-up of that project, and of others like it.\n\n## FAQ\n\n**What is structure-based drug design?**\nDesigning drug candidates from the 3D structure of the target. You read the binding pocket, place candidate molecules in it, score how they sit, then refine the ones that score well for potency and drug-like properties.\n\n**What are the best tools for structure-based drug design?**\nSchrödinger Maestro, MOE, and OpenEye each cover docking and optimization inside a full suite. Nanome adds shared 3D review over the top, in a browser or a headset, with MARA running docking, de novo binder design, and electrostatics on request.\n\n**Can I do docking without writing code?**\nYes. Describe the job to MARA in plain English and it runs the docking, co-folding, or electrostatics, then reports which tool produced the result.\n\n**Does Nanome work without a VR headset?**\nYes. There's a browser web app and a Windows desktop build, and headsets are optional: Apple Vision Pro, Pico Neo, Meta Quest, and HTC Vive Focus 3 all work.\n","2026-07-15T01:23:52.164Z","2026-09-22T16:00:10.733Z","2026-09-22T16:00:10.650Z","The best tools for structure-based drug design, from Schrödinger and MOE to Nanome and MARA for collaborative visualization and AI-run docking.","structure-based drug design, best tools for structure-based drug design, what is structure-based drug design, SBDD tools, molecular docking, de novo binder design, Nanome, MARA","the-best-tools-for-structure-based-drug-design",{"id":53,"attributes":54},80,{"title":55,"content":56,"createdAt":57,"updatedAt":58,"publishedAt":59,"date":37,"description":60,"keywords":61,"slug":62,"category":17},"The best free molecular visualization software","The best free molecular visualization software depends on what you're doing, and the short list of open tools comes down to [PyMOL](https:\u002F\u002Fpymol.org) (the open-source build), [UCSF ChimeraX](https:\u002F\u002Fwww.rbvi.ucsf.edu\u002Fchimerax\u002F), [VMD](https:\u002F\u002Fwww.ks.uiuc.edu\u002FResearch\u002Fvmd\u002F), and [Avogadro](https:\u002F\u002Favogadro.cc). Nanome fits here too. It's a collaborative molecular visualization and drug discovery platform that runs in [a browser web app and on XR headsets](https:\u002F\u002Fnanome.ai\u002Fsetup), with an AI copilot called [MARA](https:\u002F\u002Fnanome.ai\u002Fmara) built in, and its [free Starter web seat](https:\u002F\u002Fnanome.ai\u002Fpricing) opens PDB and SDF structures in a browser with no headset and no install.\n\nAll 5 have a genuinely free path. What \"free\" buys differs in each case.\n\n## What \"free\" actually means here\n\n![A researcher wearing an ultra-thin VR headset studies a ribbon-cartoon protein structure held close to the chest](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fthe_best_free_molecular_visualization_software_image_1_v4_eef934676e.png)\n\n\n\nFree splits two ways. Open-source software (the PyMOL source build, ChimeraX, VMD, Avogadro) downloads and runs on your own hardware, usually under an academic or non-commercial license. A free tier of a commercial product, like Nanome's Starter web seat, hands you a working account at no cost, with the rest of the product behind paid tiers.\n\nBoth count as free. The trade sits in different places: a local install with a deep scripting layer on one side, a browser session other people can walk into on the other.\n\n## Four open tools, and a free seat\n\n\u003Cdiv class=\"logo-row\" style=\"display:flex;flex-wrap:wrap;align-items:center;justify-content:center;gap:1.75rem;background:#fff;border-radius:12px;padding:1.25rem 1.5rem;margin:2rem 0\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fthe_best_free_molecular_visualization_software_image_4_da3f40593e.png\" alt='PyMOL' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fthe_best_free_molecular_visualization_software_image_6_4f9eb8d246.png\" alt='VMD' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fthe_best_free_molecular_visualization_software_image_7_83f950366e.png\" alt='Avogadro' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fthe_best_free_molecular_visualization_software_image_9_07bd0ccfe0.png\" alt='OpenEye' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003C\u002Fdiv>\n\n\n\n\u003Ctable class=\"table\">\n  \u003Cthead>\n    \u003Ctr>\u003Cth>Tool\u003C\u002Fth>\u003Cth>Strongest at\u003C\u002Fth>\u003Cth>What Nanome adds beside it\u003C\u002Fth>\u003C\u002Ftr>\n  \u003C\u002Fthead>\n  \u003Ctbody>\n    \u003Ctr>\u003Ctd>PyMOL (open-source build)\u003C\u002Ftd>\u003Ctd>Rendered figures for papers, Python scripting, a wide plugin ecosystem. Desktop, one person at the keyboard.\u003C\u002Ftd>\u003Ctd>Nanome reads PyMOL \u003Ccode>.pse\u003C\u002Fcode> sessions on import and wraps live multi-user review and native 3D around the same PDB and SDF coordinates.\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>UCSF ChimeraX\u003C\u002Ftd>\u003Ctd>Successor to Chimera. Density maps, very large assemblies, an analysis command line beside the GUI.\u003C\u002Ftd>\u003Ctd>Nanome opens the same coordinate files and holds them in a workspace colleagues can join from a browser tab.\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>VMD\u003C\u002Ftd>\u003Ctd>Very large systems and simulation output, scripted through Tcl and Python. A long history in simulation-heavy labs.\u003C\u002Ftd>\u003Ctd>Nanome opens the same structures and puts a group inside them at full scale.\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Avogadro\u003C\u002Ftd>\u003Ctd>Building and editing small molecules, quick geometry cleanup. Light, and quick to pick up.\u003C\u002Ftd>\u003Ctd>Nanome opens SDF ligands and fetches from PubChem, so a molecule sketched in Avogadro lands in a shared workspace.\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Nanome (free Starter seat)\u003C\u002Ftd>\u003Ctd>PDB and SDF structures in 3D in a browser tab, with nothing installed and no headset in the room.\u003C\u002Ftd>\u003Ctd>This is the free tier. Headset access and the full MARA tool library sit on the paid tiers.\u003C\u002Ftd>\u003C\u002Ftr>\n  \u003C\u002Ftbody>\n\u003C\u002Ftable>\n\nPyMOL is the usual first stop. The open-source build costs nothing, the scripting goes deep, and the renders end up in papers. Nanome imports PyMOL `.pse` sessions, while saving out stops at PDB, SDF and SMILES, so file traffic between the two runs one direction. [More on supported formats](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n\nChimeraX is the specialist for cryo-EM density maps and big assemblies. Avogadro suits sketching a small molecule and tidying its geometry. VMD sits with the simulation-heavy groups. All 4 run locally, all 4 cost nothing, and for solo desktop work they cover the job. A wider survey that takes in the paid suites as well is in [the best molecular visualization tools](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fthe-best-molecular-visualization-tools).\n\nNanome opens the coordinate files those tools write: PDB and mmCIF, SDF, MOL2, SMILES, XYZ, PQR, and PDBQT since 2.6.0. A structure crosses over without a conversion step in between.\n\n## Where Nanome's free seat comes in\n\n![Two colleagues in a bright lounge area discuss a space-filling protein structure shown on a large flat display, both pointing toward the same region of the molecule.](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fthe_best_free_molecular_visualization_software_image_2_384e519403.png)\n\n\n\nThe free Starter web seat runs in a browser tab. It opens PDB and SDF structures and fetches straight from the RCSB PDB, PubChem and DrugBank, so a structure arrives without a download and a drag.\n\nTwo things set it apart from a desktop viewer. Several people hold one structure at the same time, each reaching into the model from wherever they are, instead of one person narrating a screen share. And the workspace renders molecules in native 3D, which is what the headset tiers build on.\n\nNanome also sits beside the software a group already owns. [KNIME](https:\u002F\u002Fwww.knime.com) and Jupyter on the pipeline side, [Schrödinger LiveDesign](https:\u002F\u002Fnanome.ai\u002Fintegrations) and Cresset Flare on the modeling side, [CDD Vault](https:\u002F\u002Fwww.collaborativedrug.com) on the data side. It tends to work alongside those tools rather than take their place.\n\n## What the paid tiers add\n\nThe free web seat covers browser visualization. Two things sit above it.\n\nHeadsets are the first: a structure at full scale in native 3D, on [Apple Vision Pro, Meta Quest, HTC Vive Focus 3 or Pico Neo](https:\u002F\u002Fnanome.ai\u002Fsetup), with a Windows desktop build alongside them.\n\nNanome's AI copilot, MARA, is the second. It reaches [300+ integrated scientific tools](https:\u002F\u002Fnanome.ai\u002Fintegrations) across 26 categories, with a REST API and MCP servers underneath for scripted access. You describe the job in plain English and MARA runs it: docking with Smina or DiffDock-L, structure prediction on engines including [AlphaFold 3](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41586-024-07487-w) and [Boltz-2](https:\u002F\u002Fgithub.com\u002Fjwohlwend\u002Fboltz), electrostatics through APBS, ADMET and toxicity prediction, de novo binder design with RFdiffusion3 (beta), sequence design with ProteinMPNN, and antibody numbering and CDR definition with ANARCI. Every run names the tool it called, the inputs it took and what came back, so a reviewer can retrace it.\n\nThe antibody side of that gets a fuller treatment in [software for computational antibody design](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fsoftware-for-computational-antibody-design), and where this layer sits in a broader stack is the subject of [drug discovery software for computational chemistry](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fdrug-discovery-software-for-computational-chemistry).\n\n## Where each one lands\n\n![A researcher wearing an ultra-thin VR headset studies a solid protein surface model with a small ligand visible in its binding pocket held close to the chest](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fthe_best_free_molecular_visualization_software_image_3_v4_86dd91105e.png)\n\n\n\nFor a scripted, publication-ready still on your own machine, the open-source PyMOL build is the shortest path. For cryo-EM density work, ChimeraX. For sketching and cleaning up a small molecule, Avogadro. Nanome is the one to open when more than one person needs to be inside the structure, or when MARA should run the heavier computational tools from a plain-English request.\n\nLifeArc made the access argument in its own words. It named Nanome its externalization partner and embedded 13 Nanome experts, describing the purpose as \"democratise computational tools for all scientists\", with the aim of improving patient outcomes and shortening the time to impact. A browser seat that costs nothing is the small end of that same idea. The write-ups are collected at [nanome.ai\u002Fcase-studies](https:\u002F\u002Fnanome.ai\u002Fcase-studies), one per group.\n\n## FAQ\n\n**What is the best free molecular visualization software?**\nFor solo desktop work, the open-source PyMOL build, UCSF ChimeraX, VMD and Avogadro all cost nothing and all do the job well. For 3D viewing in a browser with colleagues in the same workspace, Nanome's free Starter web seat opens PDB and SDF structures with nothing installed.\n\n**Is Nanome free?**\nThere's a free Starter web seat for browser-based visualization of PDB and SDF structures. Headset access and the full MARA tool library are on the paid tiers.\n\n**Can I use free molecular visualization software in a browser?**\nYes. Nanome's free web seat runs in an ordinary browser tab and fetches structures from the RCSB PDB, PubChem and DrugBank, so no headset and no local install come into it.\n\n**Which file formats does Nanome open?**\nImport covers the coordinate formats: PDB (`.pdb`, `.ent`), SDF (`.sdf`, `.mol`), mmCIF (`.cif`, `.mmcif`, `.bcif`), MOL2, XYZ, PQR, SMILES, and PDBQT since 2.6.0. Session and vendor files load as well: a PyMOL `.pse`, a Maestro `.mae` or `.maegz`, a MOE `.moe`. Saving out covers less ground, a single frame as PDB, SDF or SMILES, which leaves mmCIF, MAE, MOE and PSE on the import side only. Simulation frame files and `.dx` electrostatic maps follow separate rules, and the [supported formats page](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats) carries the per-format detail.\n\n**Does Nanome replace PyMOL or ChimeraX?**\nUsually no. It opens the same PDB and SDF coordinates and connects to Schrödinger LiveDesign, Cresset Flare and CDD Vault, so most groups keep a desktop viewer for renders and scripts and bring Nanome in for the shared 3D part.\n","2026-07-15T01:23:49.912Z","2026-09-22T16:00:05.450Z","2026-09-22T16:00:05.370Z","The best free molecular visualization software, from PyMOL and ChimeraX to Nanome's free browser seat for 3D structures.","best free molecular visualization software, free molecular visualization, PyMOL, ChimeraX, VMD, Avogadro, Nanome, PDB viewer, SDF viewer","the-best-free-molecular-visualization-software",{"id":64,"attributes":65},82,{"title":66,"content":67,"createdAt":68,"updatedAt":69,"publishedAt":70,"date":37,"description":71,"keywords":72,"slug":73,"category":17},"The best molecular visualization tools in 2026","The best molecular visualization tools include [PyMOL](https:\u002F\u002Fpymol.org), [UCSF ChimeraX](https:\u002F\u002Fwww.rbvi.ucsf.edu\u002Fchimerax\u002F), [VMD](https:\u002F\u002Fwww.ks.uiuc.edu\u002FResearch\u002Fvmd\u002F), [Schrödinger](https:\u002F\u002Fwww.schrodinger.com) [Maestro](https:\u002F\u002Fwww.schrodinger.com\u002Fplatform\u002Fproducts\u002Fmaestro\u002F), [MOE](https:\u002F\u002Fwww.chemcomp.com), [BIOVIA Discovery Studio](https:\u002F\u002Fwww.3ds.com\u002Fproducts\u002Fbiovia\u002Fdiscovery-studio), [Avogadro](https:\u002F\u002Favogadro.cc), and Nanome. Most of them are single-user desktop viewers or full computational chemistry suites. Nanome is the collaborative option: a molecular visualization and drug discovery platform that runs across the web app and XR headsets, and it integrates with the suites rather than trying to replace them. An AI copilot called [MARA](https:\u002F\u002Fnanome.ai\u002Fmara) sits inside it and drives the tool runs.\n\nThat list has been stable for years, and the evergreen version of this comparison lives in [the best molecular visualization tools](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fthe-best-molecular-visualization-tools). This page is about the date on it. Here are the 4 changes that landed during 2026 and what each one does to the moment a human finally looks at a structure.\n\n## The shortlist that carried over\n\n\u003Cdiv class=\"logo-row\" style=\"display:flex;flex-wrap:wrap;align-items:center;justify-content:center;gap:1.75rem;background:#fff;border-radius:12px;padding:1.25rem 1.5rem;margin:2rem 0\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fthe_best_molecular_visualization_tools_in_2026_image_6_1ebf456527.png\" alt='VMD' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fthe_best_molecular_visualization_tools_in_2026_image_8_f3c30c3b00.png\" alt='MOE' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003C\u002Fdiv>\n\n\n\n![Two colleagues wearing ultra-thin VR headsets examine the same ribbon-cartoon protein structure floating between them](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fthe_best_molecular_visualization_tools_in_2026_image_1_v4s_630455e640.png)\n\n\n\nThe desktop names on that list are the same ones as a year ago, and each still owns the job it was written for. The movement happened on either side of them: upstream, in how many structures a model can produce in an afternoon, and downstream, in how many people can stand around one of them.\n\n\u003Ctable class=\"table\">\n  \u003Cthead>\n    \u003Ctr>\u003Cth>What moved in 2026\u003C\u002Fth>\u003Cth>The specifics\u003C\u002Fth>\u003Cth>Effect on the review step\u003C\u002Fth>\u003C\u002Ftr>\n  \u003C\u002Fthead>\n  \u003Ctbody>\n    \u003Ctr>\u003Ctd>Co-folding\u003C\u002Ftd>\u003Ctd>AlphaFold 3 held the reference position, with OpenFold3, Boltz-2, Chai-1, ESMFold and Protenix beside it, and the Boltz line ran on to BoltzMol-1 and BoltzProt-1 in June 2026\u003C\u002Ftd>\u003Ctd>Predicted complexes arrive faster than a one-at-a-time queue can clear them, and a run record has to carry the exact model name\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Generative design\u003C\u002Ftd>\u003Ctd>RFdiffusion3 (beta) open-sourced on 2026-01-14, all-atom, roughly 10x the inference speed of RFdiffusion2\u003C\u002Ftd>\u003Ctd>Binder candidates come in batches, so triage moves into 3D instead of going pose by pose\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Headsets\u003C\u002Ftd>\u003Ctd>Apple Vision Pro joined Meta Quest, Pico Neo and HTC Vive Focus 3 on the standalone side\u003C\u002Ftd>\u003Ctd>Immersive review became a slot a group can put on the calendar every week\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Nanome 2.5\u003C\u002Ftd>\u003Ctd>Frame playback, plus one live session spanning the browser web app and the headsets\u003C\u002Ftd>\u003Ctd>A colleague joins from a laptop tab with nothing to install at their end\u003C\u002Ftd>\u003C\u002Ftr>\n  \u003C\u002Ftbody>\n\u003C\u002Ftable>\n\n## Co-folding became a queue\n\n[AlphaFold 3](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41586-024-07487-w) is the model that turned co-folding into an ordinary weekly task. One naming detail is worth getting right: the [AlphaFold Protein Structure Database](https:\u002F\u002Falphafold.ebi.ac.uk) is a different resource with a similar name, and it serves the earlier AF2 predictions rather than AlphaFold 3 output.\n\nThe field kept walking after that. [Boltz-1](https:\u002F\u002Fgithub.com\u002Fjwohlwend\u002Fboltz) arrived in November 2024 for structure, Boltz-2 added binding affinity, and in June 2026 [BoltzMol-1 and BoltzProt-1](https:\u002F\u002Flabcritics.com\u002Fblog\u002F2026\u002F06\u002F17\u002Fboltzmol-1-boltzprot-1-and-the-boltz-api-ai-drug-discovery-goes-full-stack\u002F) landed, with BoltzProt-1 taking over from BoltzGen. Boltz 2.1 went closed-source and API-only. So Boltz-2 is one engine among several rather than a fixed answer, and recording which exact model produced a given complex now matters more than it did 12 months ago. Where all of these sit in the wider computational stack is mapped in [drug discovery software for computational chemistry](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fdrug-discovery-software-for-computational-chemistry).\n\nThe practical consequence is arithmetic. When a model returns plausible complexes faster than anyone can open them one at a time, the review step is the one setting the pace.\n\n## Generative design got roughly 10x faster\n\n[RFdiffusion3](https:\u002F\u002Fwww.ipd.uw.edu\u002F2025\u002F12\u002Frfdiffusion3-now-available\u002F) (beta) was open-sourced on 2026-01-14. It is all-atom, and its inference runs about 10 times faster than RFdiffusion2, which shifts the cost of de novo binder work from generation to selection.\n\nSelection is a 3D job. A designed binder with a good score can still sit slightly wrong against its epitope, and depth is the first thing a flat screen drops. ProteinMPNN writes the sequences, ANARCI numbers the variable domains and marks their CDR loops, and a person still decides which candidates go forward. That whole path has its own write-up in [software for computational antibody design](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fsoftware-for-computational-antibody-design).\n\n## Apple Vision Pro made the review a standing format\n\n![A computational chemist reviews a molecular surface structure with a binding pocket on a large curved display in a quiet office.](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fthe_best_molecular_visualization_tools_in_2026_image_2_156939e867.png)\n\n\n\nImmersive structure review has been possible for a decade. What Apple Vision Pro did in 2026 was move it from a demo to something a team can standardize on: standalone hardware, no tether, passthrough so people in the same room still see each other. Nanome runs on it, and on Meta Quest, Pico Neo and HTC Vive Focus 3.\n\nThe result shows up on calendars rather than in benchmarks. A weekly structure review in 3D is a scheduling decision once the hardware stops being a production.\n\n## What Nanome 2.5 added\n\nVersion 2.5 put playback and shared sessions into the same room. One live session now spans the browser web app and the headsets, so the person wearing a Vision Pro and the person on a laptop are holding the same molecule, in the same orientation, at the same second.\n\nThe browser half of that carries more weight than it sounds like. A biologist joining from a tab sees the pocket exactly as the person in the headset sees it, so the discussion runs over one shared object.\n\n## Where 2026 leaves the choice\n\nSame shortlist, heavier inflow. The volume arriving at the review step climbed through 2026, and the number of ways a person can join that review climbed with it. Customer write-ups are at [nanome.ai\u002Fcase-studies](https:\u002F\u002Fnanome.ai\u002Fcase-studies).\n\n## FAQ\n\n**What changed in molecular visualization during 2026?**\nFour things. AlphaFold 3 co-folding turned routine while the Boltz family moved on past Boltz-2. RFdiffusion3 (beta) was open-sourced in January, all-atom and roughly 10 times faster at inference than RFdiffusion2. Apple Vision Pro made immersive review a format teams can schedule. And Nanome 2.5 brought the browser web app and the headsets into one live session.\n\n**Is Boltz-2 the best structure prediction model right now?**\nIt is one of several. MARA runs it next to AlphaFold 3, OpenFold3, Chai-1, ESMFold and Protenix, and the Boltz family has kept moving since, with BoltzMol-1 and BoltzProt-1 arriving in June 2026. The useful habit is recording which exact model produced a given structure.\n\n**How do the main molecular visualization tools compare?**\nThe tool-by-tool version, PyMOL through Avogadro with Nanome beside them, is in [the best molecular visualization tools](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fthe-best-molecular-visualization-tools). That comparison holds in any year, and this page covers only what shifted during 2026.\n\n**Does immersive review still require a tethered headset?**\nNo. Nanome runs on standalone hardware, Apple Vision Pro, Meta Quest, Pico Neo and HTC Vive Focus 3, and in a browser tab on a laptop or a Windows desktop for anyone staying at their own machine.\n","2026-07-15T01:23:50.109Z","2026-09-22T16:00:08.981Z","2026-09-22T16:00:08.910Z","The best molecular visualization tools in 2026, from PyMOL and ChimeraX to Nanome's collaborative XR and AI copilot.","best molecular visualization tools, what tools do computational chemists use, PyMOL, UCSF ChimeraX, VMD, Schrödinger Maestro, MOE, Nanome, molecular visualization software","the-best-molecular-visualization-tools-in-2026",{"id":75,"attributes":76},78,{"title":77,"content":78,"createdAt":79,"updatedAt":80,"publishedAt":81,"date":82,"description":83,"keywords":84,"slug":85,"category":17},"The best AI tools for drug discovery","The best AI tools for drug discovery are a stack of specialized models: [AlphaFold 3](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41586-024-07487-w) and [Boltz-2](https:\u002F\u002Fgithub.com\u002Fjwohlwend\u002Fboltz) for structure prediction, RFdiffusion3 (beta) and ProteinMPNN for de novo design, docking engines, ADMET predictors, and generative chemistry. Nanome ties them together through [MARA](https:\u002F\u002Fnanome.ai\u002Fmara), an AI copilot that runs these tools from plain-English requests, shows you exactly what ran, and deploys behind your firewall.\n\nNo single model carries a project on its own, so the work runs as a chain: fold, dock, score, filter, repeat. Five categories cover most of it, and each one has more than one credible engine behind it.\n\n## Five categories, and what runs in each\n\n\u003Cdiv class=\"logo-row\" style=\"display:flex;flex-wrap:wrap;align-items:center;justify-content:center;gap:1.75rem;background:#fff;border-radius:12px;padding:1.25rem 1.5rem;margin:2rem 0\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fthe_best_ai_tools_for_drug_discovery_image_4_1cfcbd4a28.png\" alt='AlphaFold 3' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fthe_best_ai_tools_for_drug_discovery_image_6_8ce14dce4b.png\" alt='ProteinMPNN' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fthe_best_ai_tools_for_drug_discovery_image_8_edc7494425.png\" alt='OpenEye' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fthe_best_ai_tools_for_drug_discovery_image_9_2f21ccfe28.png\" alt='MOE' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003C\u002Fdiv>\n\n\n\n**Structure prediction.** A sequence goes in, a 3D fold comes out. AlphaFold 3 is DeepMind's current public model, and Boltz-2 is one of several open engines that co-fold a protein together with its ligand. OpenFold3, Chai-1, ESMFold and Protenix cover similar ground with different tradeoffs, which is why running two of them and comparing the answers is ordinary practice.\n\n**De novo design.** RFdiffusion3 draws protein backbones that don't exist yet, all-atom, with inference roughly 10 times faster than the generation before it. ProteinMPNN then writes a sequence that should fold into that shape. Between them you get a binder for a target no database has an answer for.\n\n**Docking.** A protein and a small molecule go in, and the engine predicts the pose and scores the fit. Smina and DiffDock-L are the two engines MARA calls for this. Docking is cheap and noisy enough that the working unit is hundreds of runs.\n\n**ADMET.** Absorption, distribution, metabolism, excretion, toxicity. A compound that binds beautifully and then fails on toxicity has already burned real time and money, so these predictors sit early in the funnel. eToxPred and ToxinPred handle the toxicity half.\n\n**Generative chemistry.** The model proposes molecules against a target, scores them, and takes the survivors round again. Cheminformatics routines do the bookkeeping around that loop: similarity, properties, filtering, and the record of what got proposed when.\n\nStringing those calls into one long autonomous run raises its own questions, and [agentic AI for computational chemistry](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fagentic-ai-for-computational-chemistry) works through them.\n\n## Comparison\n\n\u003Ctable class=\"table\">\n  \u003Cthead>\n    \u003Ctr>\u003Cth>Model or category\u003C\u002Fth>\u003Cth>What it does well\u003C\u002Fth>\u003Cth>What Nanome and MARA add\u003C\u002Fth>\u003C\u002Ftr>\n  \u003C\u002Fthead>\n  \u003Ctbody>\n    \u003Ctr>\u003Ctd>AlphaFold 3, Boltz-2\u003C\u002Ftd>\u003Ctd>Structure and complex prediction\u003C\u002Ftd>\u003Ctd>MARA calls either engine and colors the result by confidence; Nanome loads it in 3D\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>RFdiffusion3 (beta)\u003C\u002Ftd>\u003Ctd>De novo protein backbone generation\u003C\u002Ftd>\u003Ctd>MARA chains backbone, sequence and fold check from one request\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>ProteinMPNN, ANARCI\u003C\u002Ftd>\u003Ctd>Sequence design; antibody numbering and CDR definition\u003C\u002Ftd>\u003Ctd>MARA runs both across an antibody campaign\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Smina, DiffDock-L\u003C\u002Ftd>\u003Ctd>Pose prediction for hits\u003C\u002Ftd>\u003Ctd>MARA docks and ranks; Nanome puts the top poses in front of the team at any scale\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>APBS\u003C\u002Ftd>\u003Ctd>Electrostatics\u003C\u002Ftd>\u003Ctd>MARA solves the field and Nanome paints it onto the surface\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>ADMET and toxicity models\u003C\u002Ftd>\u003Ctd>Filtering on druglikeness and tox risk\u003C\u002Ftd>\u003Ctd>MARA scores a series before anyone commits it to synthesis\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>\u003Ca href=\"https:\u002F\u002Fwww.schrodinger.com\">Schrödinger\u003C\u002Fa>, \u003Ca href=\"https:\u002F\u002Fwww.eyesopen.com\">OpenEye\u003C\u002Fa>, \u003Ca href=\"https:\u002F\u002Fwww.chemcomp.com\">CCG MOE\u003C\u002Fa>\u003C\u002Ftd>\u003Ctd>Full comp-chem suites\u003C\u002Ftd>\u003Ctd>Nanome integrates with all three and adds a shared 3D room on top\u003C\u002Ftd>\u003C\u002Ftr>\n  \u003C\u002Ftbody>\n\u003C\u002Ftable>\n\n## Where Nanome and MARA fit\n\n![A plain-English request flows into a central hub that fans out to four specialized scientific tool icons, with a transparency log below.](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fthe_best_ai_tools_for_drug_discovery_image_2_205a49f2dc.png)\n\n\n\nNanome is a collaborative molecular visualization and drug discovery platform. It runs in [the browser, on Windows, and in XR on Meta Quest, Vive Focus 3, Pico Neo and Apple Vision Pro](https:\u002F\u002Fnanome.ai\u002Fsetup). Structures arrive by accession code from RCSB PDB, PubChem or DrugBank, or straight off a disk as PDB or SDF.\n\nNanome's copilot, MARA, reaches [300+ integrated scientific tools across 26 categories](https:\u002F\u002Fnanome.ai\u002Fintegrations). It runs [docking, co-folding, electrostatics and ADMET prediction](https:\u002F\u002Fnanome.ai\u002Fagents), numbers antibody variable domains and marks their CDR loops with ANARCI, designs sequences with ProteinMPNN, generates backbones with RFdiffusion3, and predicts folds with Boltz-2 or AlphaFold 3.\n\nGroups add their own code to that library. An in-house script, or a model a team trained on its own data, becomes a MARA tool that anyone in the org can then call by name in a sentence.\n\nA request in plain English is enough. MARA picks the tool, runs it, and hands back a record: which tool fired, the inputs it took, the artifact it produced. When a colleague or a regulator wants to know where a number came from, that record is the answer. The day-to-day rhythm of working this way gets a fuller walkthrough in [an AI copilot for drug discovery workflows](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fan-ai-copilot-for-drug-discovery-workflows).\n\nTwo properties of the setup carry most of the weight. Several people share one 3D session and study the same binding pocket at the same moment, in the browser or in a headset. And enterprise installs of Nanome and MARA sit inside your own network, as a single tenant in the cloud or on hardware you run, so an unpublished target stays where it started.\n\nInsilico Medicine worked exactly that way on COVID-19. Their AI generated 10 novel SARS-CoV-2 protease inhibitors, and the medicinal chemists reviewed those molecules in Nanome in VR before the program went any further. The work was co-authored and posted to ChemRxiv. Insilico's CEO gave the reasoning:\n\n> \"While AI can come up with novel and diverse drug-like molecules, it is important for medicinal chemists to look at these molecules closely before placing a billion-dollar, life-or-death wager. VR enables medicinal chemists to do this.\"\n>\n> Alex Zhavoronkov, CEO, Insilico Medicine\n\nThe peer-reviewed record behind the platform is under [publications](https:\u002F\u002Fnanome.ai\u002Fpublications).\n\n## When to run the model directly\n\n![Two colleagues in a bright research lounge examine a protein-ligand surface model on a large display, one pointing at the binding pocket.](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fthe_best_ai_tools_for_drug_discovery_image_3_348c9e5eef.png)\n\n\n\nOne model at volume belongs on a queue. Folding 10,000 sequences overnight, with nothing to visualize and nobody watching, is a job for AlphaFold 3 in a script, and a graphical interface adds nothing to it.\n\nWhen the deep work of a project lives in [Schrödinger Maestro](https:\u002F\u002Fwww.schrodinger.com\u002Fplatform\u002Fproducts\u002Fmaestro\u002F), MOE or [BIOVIA Discovery Studio](https:\u002F\u002Fwww.3ds.com\u002Fproducts\u002Fbiovia\u002Fdiscovery-studio), that is where it should stay. Nanome opens Maestro `.mae` and `.maegz` (the format LiveDesign hands over) and `.moe` files for viewing, and it parses the ordinary PDB, mmCIF, SDF, MOL\u002FMOL2, XYZ and PQR a pipeline already writes. Molecules come back out as PDB, SDF or SMILES, one frame at a time. [More on supported formats](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n\nNanome connects to [CDD Vault](https:\u002F\u002Fwww.collaborativedrug.com), Cresset Flare, OpenEye (Cadence) and Schrödinger LiveDesign over an open REST API and MCP servers, so the output of a run lands next to the assay record it belongs with. What other groups have built on that footing is at [nanome.ai\u002Fcase-studies](https:\u002F\u002Fnanome.ai\u002Fcase-studies).\n\n## FAQ\n\n**What are the best AI tools for drug discovery?**\nStructure prediction runs on AlphaFold 3 and Boltz-2, with OpenFold3, Chai-1, ESMFold and Protenix alongside them. De novo design runs on RFdiffusion3 and ProteinMPNN. Docking runs on Smina and DiffDock-L, and ADMET plus toxicity models filter what survives. Nanome's MARA copilot reaches 300+ of these across 26 categories and drives them from a sentence.\n\n**What are AI-powered molecular analysis tools?**\nSoftware that predicts, generates or scores molecular structure with machine learning: folding models, binder generators, docking engines, electrostatics solvers and ADMET predictors. MARA runs them, and Nanome loads what comes back into a 3D workspace a whole team can join.\n\n**Can I run these models without writing code?**\nYes. A request in ordinary English is enough. MARA selects the tool, runs it, and reports what went in and what came out, and the structures it produced land in the Nanome workspace ready to turn over.\n\n**Is my data safe with an AI copilot?**\nEnterprise deployments of Nanome and MARA sit inside your own network, as a single-tenant cloud instance or fully on-premises. Structures, sequences and the prompts you type stay in the environment they started in.\n","2026-07-15T01:23:49.712Z","2026-09-17T16:00:08.446Z","2026-09-17T16:00:08.403Z","2026-09-17","The best AI tools for drug discovery, by category, plus how Nanome and MARA run them in plain English.","best AI tools for drug discovery, AI-powered molecular analysis tools, AlphaFold, Boltz-2, RFdiffusion, ProteinMPNN, ADMET prediction, MARA, Nanome","the-best-ai-tools-for-drug-discovery",{"id":87,"attributes":88},79,{"title":89,"content":90,"createdAt":91,"updatedAt":92,"publishedAt":93,"date":82,"description":94,"keywords":95,"slug":96,"category":17},"The best cheminformatics tools for drug discovery teams","The best cheminformatics tools for drug discovery teams include [RDKit](https:\u002F\u002Fwww.rdkit.org) for scripting, [KNIME](https:\u002F\u002Fwww.knime.com) for visual pipelines, [CDD Vault](https:\u002F\u002Fwww.collaborativedrug.com) for data management, and the [Schrödinger](https:\u002F\u002Fwww.schrodinger.com) suite for modeling. Nanome sits alongside them as a collaborative molecular visualization and drug discovery platform with [an AI copilot called MARA](https:\u002F\u002Fnanome.ai\u002Fmara), which runs cheminformatics workflows from plain-English requests across [300+ integrated scientific tools](https:\u002F\u002Fnanome.ai\u002Fintegrations) and [connects directly to CDD Vault](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fcollaborative-drug-discovery-and-nanome-partnership-announcement). It works in [a browser web app, on Windows desktop, and in XR headsets](https:\u002F\u002Fnanome.ai\u002Fsetup).\n\n## What a cheminformatics stack has to do\n\n![Two colleagues wearing ultra-thin VR headsets examine the same ball-and-stick molecular structure floating between them](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fthe_best_cheminformatics_tools_for_drug_discovery_teams_image_1_v4s_d73a459442.png)\n\n\n\nCheminformatics is the part of drug discovery that turns molecules into data you can compute on. A few things fall under it:\n\n- **Property calculation.** LogP, molecular weight, hydrogen bond donors, topological polar surface area, and other descriptors you compute from a structure.\n- **Similarity search.** Fingerprints (Morgan, ECFP) and Tanimoto scores to find compounds that look like a hit.\n- **SAR analysis.** Structure-activity relationships: matched molecular pairs, R-group decomposition, activity cliffs, and how a change to a scaffold moves potency.\n- **Filtering.** Lipinski, PAINS, drug-likeness, and synthetic accessibility filters to trim a library before you spend money on it.\n- **Databases.** Querying ChEMBL, PubChem, DrugBank, and your own internal assay data.\n\nMost teams run more than one package, because none of them covers that whole list well. The 3D side of the work, the viewers and structure tools, is a separate question with its own [roundup of molecular visualization tools](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fthe-best-molecular-visualization-tools).\n\n## The common tools, and where each fits\n\n\u003Cdiv class=\"logo-row\" style=\"display:flex;flex-wrap:wrap;align-items:center;justify-content:center;gap:1.75rem;background:#fff;border-radius:12px;padding:1.25rem 1.5rem;margin:2rem 0\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fthe_best_cheminformatics_tools_for_drug_discovery_teams_image_4_7dc8dad5fe.png\" alt='RDKit' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fthe_best_cheminformatics_tools_for_drug_discovery_teams_image_5_7eb2d1ebcb.png\" alt='KNIME' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fthe_best_cheminformatics_tools_for_drug_discovery_teams_image_6_f6d3f205f2.png\" alt='CDD Vault' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fthe_best_cheminformatics_tools_for_drug_discovery_teams_image_7_5940ffdb42.png\" alt='Schrödinger' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003C\u002Fdiv>\n\n\n\n\u003Ctable class=\"table\">\n  \u003Cthead>\n    \u003Ctr>\u003Cth>Tool\u003C\u002Fth>\u003Cth>Strongest at\u003C\u002Fth>\u003Cth>How Nanome pairs with it\u003C\u002Fth>\u003C\u002Ftr>\n  \u003C\u002Fthead>\n  \u003Ctbody>\n    \u003Ctr>\u003Ctd>RDKit\u003C\u002Ftd>\u003Ctd>Open-source Python toolkit. Descriptors, fingerprints, substructure search, the engine under most pipelines\u003C\u002Ftd>\u003Ctd>MARA computes the same descriptors and similarity scores off a typed request, with no script to write first\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>KNIME\u003C\u002Ftd>\u003Ctd>Visual node-based workflows for cheminformatics and data science\u003C\u002Ftd>\u003Ctd>Nanome integrates with KNIME; MARA takes the one-off steps that never justified a node graph\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>CDD Vault\u003C\u002Ftd>\u003Ctd>Hosted registration, assay data, and SAR management\u003C\u002Ftd>\u003Ctd>Nanome connects directly to CDD Vault; MARA queries a vault and loads what comes back into a shared 3D session\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Schrödinger\u003C\u002Ftd>\u003Ctd>Full modeling suite: docking, free energy, physics-based property prediction\u003C\u002Ftd>\u003Ctd>Nanome integrates with \u003Ca href=\"https:\u002F\u002Fnanome.ai\u002Fintegrations\">Schrödinger LiveDesign\u003C\u002Fa>; MARA docks and runs electrostatics for quick exploration\u003C\u002Ftd>\u003C\u002Ftr>\n  \u003C\u002Ftbody>\n\u003C\u002Ftable>\n\nRDKit is the free foundation almost everyone builds on. KNIME turns those pieces into repeatable pipelines without much code. CDD Vault keeps compound and assay data organized and searchable. Schrödinger goes deep on physics-based modeling when the accuracy has to hold up.\n\n## How Nanome joins that stack\n\n![A flat vector diagram showing a speech bubble leading to a tool icon and then to a checklist, illustrating how a plain-English request drives an auditable cheminformatics workflow.](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fthe_best_cheminformatics_tools_for_drug_discovery_teams_image_2_2e97d182e4.png)\n\n\n\nNanome opens what those pipelines hand off. PDB and SDF files load directly, SMILES can be typed straight in, and an entry can be fetched by code from RCSB PDB or looked up as a compound in PubChem or DrugBank. SMILES comes back out too, through the rdkit-backed WorkspaceAPI\u002FMCP `export_entry` route, so a molecule round-trips between a script and a live 3D session. Two people can share that session from a browser, a Windows desktop, or a headset.\n\nMARA takes a cheminformatics step described in plain English and runs the matching tool out of a library spanning 26 categories. It calculates descriptors, computes Tanimoto similarity, applies drug-likeness and PAINS filters, scores synthetic accessibility, decomposes R-groups, mines matched molecular pairs, and flags activity cliffs. Docking, ADMET, and structure prediction sit in the same set, and there's a wider tour of those in [drug discovery software for computational chemistry](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fdrug-discovery-software-for-computational-chemistry). Teams add to it as well: a group can wrap its own script as a MARA tool and publish it internally, so a house procedure becomes something any colleague can call by name.\n\nEvery run leaves a record of the tool that fired, the inputs it took, and the numbers it returned, so a result in a report traces back to the call that produced it. Enterprise deployments can also run [on infrastructure you control](https:\u002F\u002Fnanome.ai\u002Fmara-enterprise), which tends to be a precondition when the compounds are proprietary.\n\nThe CDD Vault connection is what ties assay data back to structure. MARA can run a saved search or a similarity query against your vault and load the results into a shared session, where the chemists and the biologists read the same SAR at the same time. Reviews like that move real decisions. Nimbus Therapeutics examined AMPKβ2 in VR while the protein moved, revised a selectivity strategy it had already settled on, and came away with compounds more active on the target.\n\n## Where the other tools stay in front\n\n![A researcher at a workstation studies a cartoon-ribbon protein structure on a large monitor in a quiet modern office.](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fthe_best_cheminformatics_tools_for_drug_discovery_teams_image_3_0829056a4b.png)\n\n\n\nLarge batch pipelines that have to execute unattended every night belong in KNIME or a scheduled RDKit job. Physics-based free energy calculations at production accuracy belong in Schrödinger's suite, and the Maestro files that come out of it, `.mae` and `.maegz`, open in Nanome afterward for review (`.mae` doubles as what the LiveDesign gadget ingests). Nanome integrates with both instead of replacing them, so the usual arrangement leaves the pipeline and the modeling suite where they are and adds Nanome for the collaborative, exploratory, ask-in-plain-English part.\n\nBiologics change the tool list again, and [software for computational antibody design](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fsoftware-for-computational-antibody-design) covers that side. For what any of this looked like on live projects, the [case studies](https:\u002F\u002Fnanome.ai\u002Fcase-studies) have the specifics.\n\n## FAQ\n\n**What is the best free cheminformatics tool?**\nRDKit. It's open source, widely maintained, and does descriptors, fingerprints, similarity, and substructure search. It's also the engine underneath a lot of commercial tools. Nanome's MARA returns the same quantities from a typed request, no Python required.\n\n**Can I run cheminformatics without coding?**\nYes. KNIME gives you visual node-based pipelines. Nanome's MARA covers one-off tasks: a request like \"calculate LogP and drug-likeness for these compounds\" runs the matching tool and returns the result, along with a note of which tool handled it.\n\n**What file formats does Nanome support?**\nSupport comes in tiers, so each format below is paired with what you can do with it.\n\n- **Import, view, and edit:** PDB (`.pdb`, `.ent`), mmCIF\u002FPDBx (`.cif`, `.mmcif`, `.mcif`, `.bcif`), SDF (`.sdf`, `.sd`), MOL and MOL2, SMILES (`.smi`, or typed in), XYZ (`.xyz`), PQR (`.pqr`).\n- **Import and view:** PDBQT (`.pdbqt`, converted to PDB with charges dropped), Maestro (`.mae`, `.maegz`), MOE (`.moe`), PyMOL sessions (`.pse`).\n- **Overlay onto a model already loaded:** DX electrostatic maps (`.dx`).\n- **Export:** PDB, SDF, or SMILES, single frame. SMILES leaves through the WorkspaceAPI\u002FMCP `export_entry` route. mmCIF, MAE, MOE, and PSE are import-only.\n\nThe [file formats documentation](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats) has the rest.\n\n**Does Nanome work with CDD Vault?**\nYes. Nanome connects directly to Collaborative Drug Discovery's CDD Vault. MARA can run a saved search or a similarity query against your vault, then load what it finds into a shared 3D visualization.\n\n**What cheminformatics can MARA run?**\nProperty and descriptor calculation, Tanimoto similarity, molecular filters (drug-likeness, PAINS), synthetic accessibility scoring, R-group decomposition, matched molecular pair analysis, and activity cliff detection, among 300+ integrated tools that also cover docking, ADMET, and structure prediction.\n","2026-07-15T01:23:49.810Z","2026-09-17T16:00:10.167Z","2026-09-17T16:00:10.102Z","The best cheminformatics tools for drug discovery teams: RDKit, KNIME, CDD Vault, Schrodinger, and Nanome with MARA in plain English.","best cheminformatics tools, cheminformatics software, RDKit, KNIME, CDD Vault, drug discovery tools, molecular property calculation, SAR analysis, Nanome, MARA","the-best-cheminformatics-tools-for-drug-discovery-teams",{"id":98,"attributes":99},100,{"title":100,"content":101,"createdAt":102,"updatedAt":103,"publishedAt":104,"date":82,"description":105,"keywords":106,"slug":107,"category":17},"Spatial computing for pharmaceutical research","Spatial computing for pharmaceutical research means viewing and manipulating molecular structures as true 3D objects in immersive space, then working on them with your team as if you were standing around the same physical model. Nanome is one platform pharma teams use for this. It's a collaborative molecular visualization and drug discovery platform that runs on XR headsets, Windows desktop, and a browser web app. Nanome's AI copilot, [MARA](https:\u002F\u002Fnanome.ai\u002Fmara), runs the analysis tools.\n\nA protein has depth, and a headset hands that depth back. Someone can walk around the structure at whatever scale suits the question, put a hand into the binding pocket, and turn it while a colleague standing opposite watches the same atoms move. The headsets in play are [Meta Quest, HTC Vive Focus 3, Apple Vision Pro and Pico Neo, with a Windows app and a browser for anyone without one](https:\u002F\u002Fnanome.ai\u002Fsetup).\n\n## What spatial computing changes in R&D\n\n![A researcher wearing a slim VR headset turns a ribbon-cartoon protein structure floating at chest height in an open studio space, both hands engaged with the model.](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fspatial_computing_for_pharmaceutical_research_image_1_r3_f0b4225b1c.png)\n\n\n\nDepth carries most of the argument. On a flat screen, shading and rotation stand in for the third dimension, and the reconstruction happens in the head of whoever is looking. Give the structure real volume and real scale and that step drops out. A pocket reads as a cavity with room in it, and a bad contact sitting behind a side chain is visible on the first turn.\n\nPresence is the second change. Two scientists on opposite coasts stand in one room and argue about a docked pose that sits in front of both of them, which is the working pattern [collaborative drug discovery software for remote teams](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fcollaborative-drug-discovery-software-for-remote-teams) goes through in detail.\n\nImmersive space is doing interface work here. It puts a person at the scale the chemistry happens on, which is where molecular design has been short for a while.\n\nOn an org chart, spatial computing turns up in the review meeting, where the shared view is the object a chemist, a biologist and a modeler can all point at, and [what a modern molecular presentation looks like](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fwhat-a-modern-molecular-presentation-looks-like) covers how that session gets built.\n\n## Where the tools come in\n\nA session opens on the real thing. A structure arrives by accession code from RCSB PDB, PubChem, DrugBank, ChEMBL or UniProt, or straight off a disk. Seeing it settles some questions; whether a compound binds, folds, or trips a toxicity flag takes a calculation.\n\nThat half goes to MARA, Nanome's AI copilot, which carries [300+ integrated scientific tools](https:\u002F\u002Fnanome.ai\u002Fintegrations) spread over 26 categories. Ask for one in plain English and the result lands back in the same 3D scene. With a headset on, the request can be spoken out loud instead of typed, which keeps both hands on the molecule.\n\nSome of the ground it covers: docking with Smina or DiffDock-L; folding and co-folding through [AlphaFold 3](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41586-024-07487-w), [Boltz-2](https:\u002F\u002Fgithub.com\u002Fjwohlwend\u002Fboltz), OpenFold3 and Chai-1; electrostatics with APBS; ADMET and toxicity models; ProteinMPNN for sequence design, with ANARCI numbering antibody variable domains and marking their CDR loops; and RFdiffusion3 (beta) for de novo binders.\n\nEach run leaves a record of the tool, what went into it and what came out, so a number on a slide can be walked back to the job behind it. Once a pose is seated, [how to analyze protein-ligand interactions](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fhow-to-analyze-protein-ligand-interactions) covers the measurements that follow.\n\n## How Nanome compares to other visualization tools\n\n\u003Cdiv class=\"logo-row\" style=\"display:flex;flex-wrap:wrap;align-items:center;justify-content:center;gap:1.75rem;background:#fff;border-radius:12px;padding:1.25rem 1.5rem;margin:2rem 0\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fspatial_computing_for_pharmaceutical_research_image_4_7e4d45002b.png\" alt='PyMOL' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fspatial_computing_for_pharmaceutical_research_image_5_08ccd52555.png\" alt='UCSF ChimeraX' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fspatial_computing_for_pharmaceutical_research_image_6_8767392b4b.png\" alt='VMD' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fspatial_computing_for_pharmaceutical_research_image_7_9de7e5b8bc.png\" alt='Schrödinger Maestro' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fspatial_computing_for_pharmaceutical_research_image_8_5776041dfb.png\" alt='MOE' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003C\u002Fdiv>\n\n\n\n\u003Ctable class=\"table\">\n  \u003Cthead>\n    \u003Ctr>\u003Cth>Tool\u003C\u002Fth>\u003Cth>Its strength\u003C\u002Fth>\u003Cth>What immersive space adds\u003C\u002Fth>\u003C\u002Ftr>\n  \u003C\u002Fthead>\n  \u003Ctbody>\n    \u003Ctr>\u003Ctd>\u003Ca href=\"https:\u002F\u002Fpymol.org\">PyMOL\u003C\u002Fa>, \u003Ca href=\"https:\u002F\u002Fwww.rbvi.ucsf.edu\u002Fchimerax\u002F\">ChimeraX\u003C\u002Fa>, \u003Ca href=\"https:\u002F\u002Fwww.ks.uiuc.edu\u002FResearch\u002Fvmd\u002F\">VMD\u003C\u002Fa>\u003C\u002Ftd>\u003Ctd>Deep scripting and rendering on the desktop, one operator at a time\u003C\u002Ftd>\u003Ctd>A structure at body scale that a group holds together\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>\u003Ca href=\"https:\u002F\u002Fwww.schrodinger.com\u002Fplatform\u002Fproducts\u002Fmaestro\u002F\">Schrödinger Maestro\u003C\u002Fa>, \u003Ca href=\"https:\u002F\u002Fwww.chemcomp.com\">MOE\u003C\u002Fa>, \u003Ca href=\"https:\u002F\u002Fwww.3ds.com\u002Fproducts\u002Fbiovia\u002Fdiscovery-studio\">BIOVIA Discovery Studio\u003C\u002Fa>\u003C\u002Ftd>\u003Ctd>Full comp-chem suites for the modeling itself\u003C\u002Ftd>\u003Ctd>Nanome connects to several of them (\u003Ca href=\"https:\u002F\u002Fnanome.ai\u002Fintegrations\">Schrödinger LiveDesign\u003C\u002Fa>) and puts their output in a room\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Web-only molecular viewers\u003C\u002Ftd>\u003Ctd>Fast structure lookups in a browser tab\u003C\u002Ftd>\u003Ctd>Nanome's web app covers that, then carries the same session into a headset\u003C\u002Ftd>\u003C\u002Ftr>\n  \u003C\u002Ftbody>\n\u003C\u002Ftable>\n\nNanome imports PyMOL `.pse` sessions, along with the structure files those desktop tools write every day: PDB, mmCIF, SDF, MOL and MOL2, XYZ and PQR. From the Schrödinger side it takes Maestro `.mae` and `.maegz`, the format LiveDesign ingests, and it reads `.moe` files from MOE. [More on supported formats](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n\nWhatever pipeline a group already runs can stay put. Nanome connects to [Schrödinger](https:\u002F\u002Fwww.schrodinger.com) LiveDesign, Cresset Flare, [OpenEye \u002F Cadence](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fview-your-openeyecadence-data-with-fresh-eyes-using-nanome-xr), [CDD Vault](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fcollaborative-drug-discovery-and-nanome-partnership-announcement), [KNIME](https:\u002F\u002Fwww.knime.com), Jupyter, and the [OpenFold Consortium](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fnanome-joins-the-openfold-consortium).\n\n## Real work done this way\n\n![Two colleagues wearing ultra-thin VR headsets examine the same space-filling protein model floating between them](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fspatial_computing_for_pharmaceutical_research_image_2_v4s_899f025ad1.png)\n\n\n\nNimbus Therapeutics had settled on a selectivity strategy for the AMPKβ2 enzyme. Seeing the protein in motion in VR, the team spotted a better synthetic vector and changed course, and the compounds that came out of the new plan were more active on the target.\n\nNimbus works against a lead-optimization cycle that runs 12 to 18 months, and a course correction found in a review arrives before the chemistry gets made.\n\n## Deployment and security\n\n![A flat vector diagram shows a molecular structure and an AI model both contained inside a single building outline, with no data crossing the perimeter wall.](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fspatial_computing_for_pharmaceutical_research_image_3_bc102c949c.png)\n\n\n\nFor internal tooling there's an open REST API, MCP servers, and a Nanome Claude Code Skill.\n\n## What a desktop tool still does better\n\nScripting a publication figure is quicker in PyMOL or ChimeraX, and for solo rendering work that's usually the better route. A calculation that already lives in a Schrödinger or MOE workflow can stay exactly where it is; Nanome takes the output and puts it in front of the group in 3D.\n\n[nanome.ai\u002Fcase-studies](https:\u002F\u002Fnanome.ai\u002Fcase-studies) collects write-ups of how different groups run this in practice.\n\n## FAQ\n\n**What is spatial computing for pharmaceutical research?**\nUsing immersive 3D and XR headsets to view, manipulate, and collaborate on molecular structures as real spatial objects. Nanome is one platform pharma teams use for it, on Pico Neo, HTC Vive Focus 3, Meta Quest and Apple Vision Pro, and on Windows desktop and in a browser.\n\n**Do I need a headset to use Nanome?**\nNo. The browser web app runs without one, and there's a Windows desktop app as well. Full immersion needs a headset; loading structures and running MARA tools does not.\n\n**Does Apple Vision Pro work with Nanome?**\nYes. Apple Vision Pro is supported, and so are the Quest, Focus 3 and Neo headsets.\n\n**What file formats can Nanome open?**\nStructures first: PDB (`.pdb`, `.ent`), SDF, MOL and MOL2, mmCIF (`.cif`, `.mmcif`, `.bcif`), SMILES, XYZ, PQR, and PDBQT, which comes in as PDB with its charges dropped. Vendor and session files load too: Maestro `.mae` and `.maegz`, MOE `.moe`, and PyMOL `.pse`. An electrostatic map (`.dx`) overlays a structure that's already open. Coming back out, a molecule saves as PDB, SDF or SMILES, and only the frame on screen travels with it, which leaves mmCIF, MAE, MOE and PSE on the import side. The full table sits in [the format docs](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n\n**Can spatial computing run real drug discovery calculations, or is it only visualization?**\nBoth. Through MARA, Nanome reaches 300+ tools covering docking, co-folding, ADMET prediction, binder design and structure prediction, with results returned into the same 3D scene and a record of every job that ran.\n","2026-07-15T01:23:52.068Z","2026-09-17T16:00:06.711Z","2026-09-17T16:00:06.651Z","Spatial computing for pharmaceutical research: Nanome runs immersive 3D structure review and MARA tools across headsets and the web.","spatial computing for pharmaceutical research, pharma spatial computing, Apple Vision Pro drug discovery, immersive molecular visualization, Nanome, MARA, XR drug discovery","spatial-computing-for-pharmaceutical-research",{"id":109,"attributes":110},76,{"title":111,"content":112,"createdAt":113,"updatedAt":114,"publishedAt":115,"date":82,"description":116,"keywords":117,"slug":118,"category":17},"Running AlphaFold and Boltz-2 in a molecular workflow","Nanome is a collaborative molecular visualization and drug discovery platform that runs across [a browser web app, XR headsets, and Windows desktop](https:\u002F\u002Fnanome.ai\u002Fsetup). It also ships [an AI copilot called MARA](https:\u002F\u002Fnanome.ai\u002Fmara). To run [AlphaFold 3](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41586-024-07487-w) or [Boltz-2](https:\u002F\u002Fgithub.com\u002Fjwohlwend\u002Fboltz) inside a workflow, you ask MARA for the prediction in plain English. MARA calls the model and reports the confidence scores, then Nanome opens the predicted structure in real-time 3D or in a headset, so a sequence turns into something you can walk around without wiring two programs together yourself. Both engines sit in MARA's [300+ integrated scientific tools](https:\u002F\u002Fnanome.ai\u002Fintegrations), and that catalogue takes additions: a group can wrap its own prediction step as a MARA tool and publish it internally, so an in-house method becomes a sentence anyone on the team can say.\n\n## The shape of a prediction-to-visualization workflow\n\n![A researcher wearing an ultra-thin VR headset studies a ribbon-cartoon protein structure held close to the chest](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Frunning_alphafold_and_boltz_2_in_a_molecular_workflow_image_1_v4_a88a799b6a.png)\n\n\n\nA prediction pays off once you can grade it and then look at it. Three steps run back to back:\n\n1. **Predict.** Fold a sequence with AlphaFold 3, or co-fold a protein and a ligand together with Boltz-2 to get the bound complex.\n2. **Assess confidence.** Read the per-residue and per-interface scores, so the parts of the model worth acting on are marked out.\n3. **Visualize and analyze.** Open the structure in 3D or XR, color it by confidence, and run pockets, interactions or electrostatics on top of it.\n\nEach of those steps has strong tools behind it. The cost sits in the seams between them: an export here, a format conversion there, a confidence file that lives somewhere apart from the coordinates. MARA runs the three in sequence, and each run leaves a legible trail: the tool it called, the inputs it passed, the artifact it produced. The broader shape of that orchestration, across docking and design as well as folding, gets walked through in [an AI copilot for drug discovery workflows](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fan-ai-copilot-for-drug-discovery-workflows).\n\n## Step 1: predict a structure\n\nFor a plain fold, you give MARA a sequence and ask for AlphaFold 3. For a protein plus a small molecule, Boltz-2 co-folds the two and predicts how the ligand sits in the pocket, which helps when no crystal structure of the complex exists.\n\nBoltz-2 is one engine among several here. OpenFold3, Chai-1, ESMFold and Protenix are separate calls in the same library, so a second opinion on the same sequence is one more request.\n\nThe co-folding path is written down rather than left to be reverse-engineered: Nanome's [Boltz-2 tutorial](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fsetting-up-boltz-2-configuration-files-and-analysis-with-nanome-ai) walks through the configuration files and the analysis of what comes back.\n\n## Step 2: assess confidence\n\nAlphaFold 3 and Boltz-2 both return per-residue confidence (pLDDT), and complexes come back with interface and alignment error scores on top of that.\n\nColoring the structure by those scores is the useful move. High-confidence regions render one way and low-confidence loops another, so the solid parts and the guesswork separate visually. MARA runs the coloring and reports the numbers next to it. When the protein-ligand interface scores badly, that surfaces before a series gets built on the pose.\n\n## Step 3: visualize and analyze in 3D and XR\n\n![Two colleagues wearing ultra-thin VR headsets examine the same solid protein surface model with a small ligand visible in its binding pocket floating between them](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Frunning_alphafold_and_boltz_2_in_a_molecular_workflow_image_2_v4s_86e2fea29d.png)\n\n\n\nWith the structure predicted and scored, Nanome opens it in the workspace. You turn it over in real-time 3D in the browser, or step inside it on a Meta Quest or an Apple Vision Pro. Several people can stand in the same predicted model at once and point at the same residue, so a shaky loop gets argued about in place.\n\nThe session keeps going from there: detect pockets, map non-covalent interactions, [compute electrostatics with APBS](https:\u002F\u002Fnanome.ai\u002Fagents), or run ADMET on the bound ligand. The prediction feeds the analysis with no export and re-import in between. Chaining several of those calls into one longer autonomous run is a subject of its own, covered in [agentic AI for computational chemistry](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fagentic-ai-for-computational-chemistry).\n\n## The workflow at a glance\n\n\u003Ctable class=\"table\">\n  \u003Cthead>\n    \u003Ctr>\u003Cth>Step\u003C\u002Fth>\u003Cth>What runs\u003C\u002Fth>\u003Cth>What you get\u003C\u002Fth>\u003Cth>Where Nanome fits\u003C\u002Fth>\u003C\u002Ftr>\n  \u003C\u002Fthead>\n  \u003Ctbody>\n    \u003Ctr>\u003Ctd>Predict a fold\u003C\u002Ftd>\u003Ctd>AlphaFold 3 via MARA\u003C\u002Ftd>\u003Ctd>A predicted protein structure\u003C\u002Ftd>\u003Ctd>Called from an ordinary sentence, then opened in the workspace\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Co-fold a complex\u003C\u002Ftd>\u003Ctd>Boltz-2 via MARA\u003C\u002Ftd>\u003Ctd>Protein plus ligand, predicted bound\u003C\u002Ftd>\u003Ctd>Configuration and analysis written up in Nanome's Boltz-2 tutorial\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Compare engines\u003C\u002Ftd>\u003Ctd>OpenFold3, Chai-1, ESMFold, Protenix\u003C\u002Ftd>\u003Ctd>A second opinion on the same sequence\u003C\u002Ftd>\u003Ctd>Separate calls in the same tool library\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Assess confidence\u003C\u002Ftd>\u003Ctd>pLDDT and interface scores\u003C\u002Ftd>\u003Ctd>A graded model you can trust selectively\u003C\u002Ftd>\u003Ctd>MARA colors the structure by confidence and reports the numbers\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Visualize\u003C\u002Ftd>\u003Ctd>Nanome web app and XR\u003C\u002Ftd>\u003Ctd>The structure in real-time 3D or immersive XR\u003C\u002Ftd>\u003Ctd>Several people in one session, on a headset or in a browser\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Analyze\u003C\u002Ftd>\u003Ctd>Pockets, interactions, APBS, ADMET\u003C\u002Ftd>\u003Ctd>Downstream results on the predicted structure\u003C\u002Ftd>\u003Ctd>Same session, no export-import between tools\u003C\u002Ftd>\u003C\u002Ftr>\n  \u003C\u002Ftbody>\n\u003C\u002Ftable>\n\n## When a command line or a comp-chem suite fits better\n\n\u003Cdiv class=\"logo-row\" style=\"display:flex;flex-wrap:wrap;align-items:center;justify-content:center;gap:1.75rem;background:#fff;border-radius:12px;padding:1.25rem 1.5rem;margin:2rem 0\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Frunning_alphafold_and_boltz_2_in_a_molecular_workflow_image_5_15e0da633d.png\" alt='MOE' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Frunning_alphafold_and_boltz_2_in_a_molecular_workflow_image_6_b64336d91b.png\" alt='AlphaFold 3' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003C\u002Fdiv>\n\n\n\n![A researcher studies a space-filling protein structure on a large monitor in a quiet, uncluttered workspace.](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Frunning_alphafold_and_boltz_2_in_a_molecular_workflow_image_3_44fa09026f.png)\n\n\n\nFolding thousands of sequences as a raw batch job on a cluster, with nobody opening the results interactively, is a job for a command-line pipeline. And when a specialized prediction workflow already lives inside [Schrödinger](https:\u002F\u002Fwww.schrodinger.com) [Maestro](https:\u002F\u002Fwww.schrodinger.com\u002Fplatform\u002Fproducts\u002Fmaestro\u002F) or [MOE](https:\u002F\u002Fwww.chemcomp.com), that suite stays the right home for it. Nanome integrates with several of them, [Schrödinger LiveDesign](https:\u002F\u002Fnanome.ai\u002Fintegrations) and CCG MOE included, so most groups run both. Prepared work opens for review without a rebuild: Nanome imports Maestro `.mae` and `.maegz` (the format LiveDesign ingests) and reads `.moe`, all of it import-only. [More on supported formats](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n\nThe case for running a prediction this way is specific. The fold, the confidence read and the 3D inspection happen in one plain-English pass, and what lands at the end is a structure a whole group can stand around and argue about.\n\nInsilico Medicine generated 10 novel SARS-CoV-2 protease inhibitors with AI, then reviewed them in Nanome in VR before committing anything to synthesis, and the work was co-authored and posted to ChemRxiv. CEO Alex Zhavoronkov put the reason for that review step plainly: medicinal chemists have to look closely at an AI-generated molecule \"before placing a billion-dollar, life-or-death wager.\" Nanome's [publications](https:\u002F\u002Fnanome.ai\u002Fpublications) page collects the peer-reviewed work, and the customer accounts are in the [case studies](https:\u002F\u002Fnanome.ai\u002Fcase-studies).\n\n## FAQ\n\n**How do I run AlphaFold 3 in a workflow?**\nYou ask MARA for a fold in plain English and give it a sequence. MARA calls AlphaFold 3 through its integrated tools and reports the pLDDT confidence, and Nanome opens the predicted structure in the web app or in XR for inspection.\n\n**What does a Boltz-2 structure prediction workflow look like?**\nYou point MARA at a protein and a ligand, and Boltz-2 co-folds them into a predicted bound complex. Nanome's Boltz-2 tutorial covers the configuration files and the analysis, MARA colors the output by confidence, and the structure opens in 3D or XR.\n\n**How do I predict a structure and then visualize it?**\nThe prediction runs through MARA with AlphaFold 3 or Boltz-2, the confidence scores come back with it, and Nanome opens the result in real-time 3D or immersive XR. Pockets, interactions and electrostatics run on the predicted model in the same session.\n\n**How do I know whether a predicted structure is trustworthy?**\nThe confidence scores answer that. AlphaFold 3 and Boltz-2 both return per-residue pLDDT, and complexes get interface and alignment error scores. MARA colors the structure by confidence, so the solid regions and the shaky ones separate before anything gets built on the model.\n","2026-07-15T01:23:49.496Z","2026-09-17T16:00:04.956Z","2026-09-17T16:00:04.889Z","How to run AlphaFold in a workflow with Boltz-2 structure prediction, then assess confidence and visualize the result in 3D and XR with Nanome and MARA.","how to run AlphaFold in a workflow, Boltz-2 structure prediction workflow, predict a structure then visualize it, AlphaFold, Boltz-2, structure prediction, confidence assessment, pLDDT, Nanome, MARA","running-alphafold-and-boltz-2-in-a-molecular-workflow",{"id":120,"attributes":121},97,{"title":122,"content":123,"createdAt":124,"updatedAt":125,"publishedAt":126,"date":127,"description":128,"keywords":129,"slug":130,"category":17},"Natural language interfaces for molecular modeling","Nanome is a collaborative molecular visualization and drug discovery platform that runs in a browser web app and on XR headsets. Inside it sits [MARA](https:\u002F\u002Fnanome.ai\u002Fmara), an AI copilot. With MARA, you type or speak a request in plain English (\"dock this ligand into the pocket,\" \"predict ADMET for these 12 compounds\"), and it picks the right tool, runs it, and shows you exactly what ran, what went in, and what came out. The structures load and light up in 3D, on your desktop or inside a headset.\n\nThat's the shift. For decades, molecular modeling meant memorizing menus, chaining scripts, and knowing which flag did what. MARA lets you describe the outcome you want and handles the plumbing underneath.\n\n## How a plain-English request becomes a real tool run\n\n![A researcher wearing an ultra-thin VR headset studies a solid protein surface model with a visible binding pocket held close to the chest](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fnatural_language_interfaces_for_molecular_modeling_image_1_v4_0ee48b0868.png)\n\n\n\nMARA sits on top of [300+ integrated scientific tools across 26 categories](https:\u002F\u002Fnanome.ai\u002Fintegrations). A request maps onto one of those tools, picks up inputs from whatever structures are open, and runs the actual computation. Docking really docks. Prediction really predicts.\n\nThe part that matters for trust: MARA shows its work. Every run lists the tool it called, the exact inputs, and the raw outputs. Docking scores, binding residues, confidence values, all of it is there to check. Nothing hides behind a chat bubble. How the copilot is put together underneath is a subject of its own, covered in [an AI copilot for drug discovery workflows](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fan-ai-copilot-for-drug-discovery-workflows).\n\nThen the result goes to the viewer. A docked pose, a predicted fold, an electrostatics surface: all of it renders in the same 3D workspace you can walk around in XR or spin in a browser tab.\n\n## Example requests and the tools they trigger\n\n\u003Ctable class=\"table\">\n  \u003Cthead>\n    \u003Ctr>\u003Cth>You type or say\u003C\u002Fth>\u003Cth>What MARA runs\u003C\u002Fth>\u003Cth>What you get back\u003C\u002Fth>\u003C\u002Ftr>\n  \u003C\u002Fthead>\n  \u003Ctbody>\n    \u003Ctr>\u003Ctd>\"Dock this ligand into the ATP pocket\"\u003C\u002Ftd>\u003Ctd>Smina or DiffDock-L docking\u003C\u002Ftd>\u003Ctd>Scored poses in the pocket, viewable in 3D\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>\"Co-fold this peptide with the receptor\"\u003C\u002Ftd>\u003Ctd>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fjwohlwend\u002Fboltz\">Boltz-2\u003C\u002Fa> \u002F \u003Ca href=\"https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41586-024-07487-w\">AlphaFold 3\u003C\u002Fa> co-folding\u003C\u002Ftd>\u003Ctd>A predicted complex with confidence coloring\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>\"Predict ADMET for these compounds\"\u003C\u002Ftd>\u003Ctd>ADMET prediction\u003C\u002Ftd>\u003Ctd>Property tables per molecule\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>\"Show the electrostatic surface\"\u003C\u002Ftd>\u003Ctd>APBS electrostatics\u003C\u002Ftd>\u003Ctd>A colored potential map on the protein surface\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>\"Design 20 binders for this epitope\"\u003C\u002Ftd>\u003Ctd>RFdiffusion3 (beta) de novo binder design\u003C\u002Ftd>\u003Ctd>Candidate backbones ready to inspect\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>\"Number the CDR loops on this antibody\"\u003C\u002Ftd>\u003Ctd>ANARCI numbering and CDR definition\u003C\u002Ftd>\u003Ctd>Annotated antibody regions\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>\"Set up 4 scenes showing each pose from the same angle\"\u003C\u002Ftd>\u003Ctd>Scene creation in the workspace\u003C\u002Ftd>\u003Ctd>Four saved 3D views, each holding its own camera\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>\"Make a slide from this scene\"\u003C\u002Ftd>\u003Ctd>PowerPoint slide generation\u003C\u002Ftd>\u003Ctd>A slide built from the live structure\u003C\u002Ftd>\u003C\u002Ftr>\n  \u003C\u002Ftbody>\n\u003C\u002Ftable>\n\nThe last two rows cover what happens after the run. Scenes are saved 3D views of the same structure, each with its own angle, and a colleague can step through them on the web or in a headset, which is roughly [what a modern molecular presentation looks like](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fwhat-a-modern-molecular-presentation-looks-like).\n\nStarting structures come from RCSB PDB, PubChem, or DrugBank, or from files you already have. From there the request is just a sentence.\n\nThe catalogue grows from your side too. Teams write their own MARA tools and publish them to colleagues, so an in-house method becomes a sentence anyone in the group can say. Stringing several of those runs into one instruction is where this turns [agentic](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fagentic-ai-for-computational-chemistry).\n\n## How this sits alongside established modeling software\n\n\u003Cdiv class=\"logo-row\" style=\"display:flex;flex-wrap:wrap;align-items:center;justify-content:center;gap:1.75rem;background:#fff;border-radius:12px;padding:1.25rem 1.5rem;margin:2rem 0\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fnatural_language_interfaces_for_molecular_modeling_image_4_31dfa21e0f.png\" alt='PyMOL' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fnatural_language_interfaces_for_molecular_modeling_image_6_bd412b6913.png\" alt='VMD' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fnatural_language_interfaces_for_molecular_modeling_image_7_11ca634958.png\" alt='Schrödinger Maestro' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fnatural_language_interfaces_for_molecular_modeling_image_8_7d8300b8e8.png\" alt='MOE' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003C\u002Fdiv>\n\n\n\n![A flat diagram showing multiple specialist tool icons converging through a single interface into one unified molecular result.](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fnatural_language_interfaces_for_molecular_modeling_image_3_a1b54d5dc9.png)\n\n\n\nPlenty of software does molecular modeling well, and most of it is driven through a GUI or a scripting layer. A typed or spoken sentence is a third way in.\n\n\u003Ctable class=\"table\">\n  \u003Cthead>\n    \u003Ctr>\u003Cth>Software\u003C\u002Fth>\u003Cth>Its strength\u003C\u002Fth>\u003Cth>What Nanome adds\u003C\u002Fth>\u003C\u002Ftr>\n  \u003C\u002Fthead>\n  \u003Ctbody>\n    \u003Ctr>\u003Ctd>\u003Ca href=\"https:\u002F\u002Fpymol.org\">PyMOL\u003C\u002Fa>, \u003Ca href=\"https:\u002F\u002Fwww.rbvi.ucsf.edu\u002Fchimerax\u002F\">ChimeraX\u003C\u002Fa>, \u003Ca href=\"https:\u002F\u002Fwww.ks.uiuc.edu\u002FResearch\u002Fvmd\u002F\">VMD\u003C\u002Fa>\u003C\u002Ftd>\u003Ctd>Precise desktop viewing and scripting\u003C\u002Ftd>\u003Ctd>Spoken or typed tool runs, plus multiplayer 3D and XR over the same structures\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>\u003Ca href=\"https:\u002F\u002Fwww.schrodinger.com\u002Fplatform\u002Fproducts\u002Fmaestro\u002F\">Schrödinger Maestro\u003C\u002Fa>, \u003Ca href=\"https:\u002F\u002Fwww.chemcomp.com\">MOE\u003C\u002Fa>, \u003Ca href=\"https:\u002F\u002Fwww.3ds.com\u002Fproducts\u002Fbiovia\u002Fdiscovery-studio\">Discovery Studio\u003C\u002Fa>\u003C\u002Ftd>\u003Ctd>Deep desktop comp-chem suites\u003C\u002Ftd>\u003Ctd>MARA runs many of the same analyses from a typed request, and Nanome integrates with several of these suites rather than replacing them\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Notebook-driven pipelines (\u003Ca href=\"https:\u002F\u002Fwww.knime.com\">KNIME\u003C\u002Fa>, Jupyter)\u003C\u002Ftd>\u003Ctd>Reproducible scripted workflows\u003C\u002Ftd>\u003Ctd>An API and MCP servers, so a request and a pipeline work on the same structures\u003C\u002Ftd>\u003C\u002Ftr>\n  \u003C\u002Ftbody>\n\u003C\u002Ftable>\n\nThe handoff happens at the file level. Nanome reads PyMOL `.pse` sessions, Maestro `.mae` and `.maegz` (the format LiveDesign ingests), and MOE `.moe`, all of them import-only. Structures arrive as PDB, mmCIF, SDF, MOL and MOL2, XYZ, PQR, or PDBQT, and export is PDB, SDF, or SMILES, one frame at a time. The full tiers per format sit in the [supported formats docs](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n\nNanome connects outward as well: [CDD Vault](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fcollaborative-drug-discovery-and-nanome-partnership-announcement), [Schrödinger LiveDesign](https:\u002F\u002Fnanome.ai\u002Fintegrations), CCG MOE, [OpenEye\u002FCadence](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fview-your-openeyecadence-data-with-fresh-eyes-using-nanome-xr), Cresset Flare, the [OpenFold Consortium](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fnanome-joins-the-openfold-consortium), plus KNIME and Jupyter.\n\nA plain-English layer buys less when a group already scripts one desktop suite fluently and spends the whole day inside it. It pays off when the work crosses many tools, when colleagues need to stand in the same 3D scene, or when picking the tool is itself the slow step.\n\n## Where the 3D and the people come in\n\n![Two colleagues wearing ultra-thin VR headsets examine the same ribbon-cartoon protein structure floating between them](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fnatural_language_interfaces_for_molecular_modeling_image_3_v4s_67b5234c40.png)\n\n\n\nReading a docking score is one thing. Standing next to the pocket with a colleague and watching the pose settle is another.\n\nThe same session opens on [Windows desktop, in a browser tab, or in a headset](https:\u002F\u002Fnanome.ai\u002Fsetup): Meta Quest, Apple Vision Pro, HTC Vive Focus 3, Pico Neo. Several people can hold one molecular scene between them at once, so a MARA result becomes something the whole group can walk around at full size.\n\nIn a headset there's no keyboard within reach. You can say the request to MARA instead and keep both hands on the molecule while it runs.\n\nProf. Radić's group at UC San Diego described the loop plainly: \"We use minimization... we can dock it to a macromolecule using Smina Docking... we can form covalent conjugates, we can minimize, dock it and score it in real time.\" Smina is still one of the docking engines a typed request reaches. More stories at [nanome.ai\u002Fcase-studies](https:\u002F\u002Fnanome.ai\u002Fcase-studies).\n\n## FAQ\n\n**What is natural language molecular modeling?**\nDescribing a modeling task in ordinary words and letting software pick and run the right computational tool. With Nanome's MARA, \"dock this ligand\" or \"predict the fold\" triggers the real docking or structure-prediction run, then shows the inputs and the outputs.\n\n**Can I really talk to my molecular data?**\nYes. You load a structure from RCSB PDB, PubChem, DrugBank, or your own files, then type or say what you want done. MARA maps the request onto one of 300+ tools and runs it, with the results rendered in 3D.\n\n**Are these plain-English drug discovery tools accurate, or just chat?**\nThe tools are the real thing: Smina and DiffDock-L docking, Boltz-2 and AlphaFold 3 co-folding, APBS electrostatics, ADMET prediction, ANARCI antibody numbering, RFdiffusion3 (beta) binder design, and cheminformatics. MARA reports the tool that ran, the exact inputs, and the raw outputs, so any result can be checked line by line.\n\n**Do I need a VR headset?**\nNo. Nanome opens in a browser tab and on Windows desktop. Headsets add shared immersive 3D (Meta Quest, Apple Vision Pro, HTC Vive Focus 3, Pico Neo) and stay optional.\n\n**Where do my structures go when MARA runs a tool?**\nFor regulated work, Nanome deploys [inside your own network, as a single-tenant cloud or on-prem](https:\u002F\u002Fnanome.ai\u002Fmara-enterprise), so structures and results stay in an environment your team controls.\n","2026-07-15T01:23:51.720Z","2026-09-15T16:00:07.690Z","2026-09-15T16:00:07.626Z","2026-09-15","Natural language molecular modeling with Nanome and MARA: type a plain-English request, watch the real tool run, and see results in 3D or XR.","natural language molecular modeling, talk to your molecular data, plain English drug discovery tools, MARA, Nanome, AI copilot molecular modeling","natural-language-interfaces-for-molecular-modeling",{"id":132,"attributes":133},89,{"title":134,"content":135,"createdAt":136,"updatedAt":137,"publishedAt":138,"date":127,"description":139,"keywords":140,"slug":141,"category":17},"Software for visualizing molecular dynamics trajectories","Nanome is a collaborative molecular visualization and drug discovery platform. It works on XR headsets and in a browser web app. An [AI copilot called MARA](https:\u002F\u002Fnanome.ai\u002Fmara) sits inside it: Nanome plays back molecular dynamics trajectories frame by frame, and MARA analyzes them. You load your trajectory, scrub through the frames, and ask MARA for RMSD, pairwise distances, or a representative energy-minimum frame in plain English. For most MD work the well-known desktop tools are [VMD](https:\u002F\u002Fwww.ks.uiuc.edu\u002FResearch\u002Fvmd\u002F) and [PyMOL](https:\u002F\u002Fpymol.org), and Nanome adds immersive 3D playback plus a copilot that runs the analysis for you.\n\nTrajectories come in the way the simulation wrote them. A `.gro` file loads on its own. Frames in `.xtc`, `.trr` or `.dcd` attach to a model already open in the workspace, and their atom count has to equal the host model's. Either route runs on the frame-trajectory substrate, which carries a 2000-frame ceiling per model. [More on supported formats](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n\n## What trajectory visualization actually needs\n\n![A researcher wearing an ultra-thin VR headset studies a ribbon-cartoon protein structure held close to the chest](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fsoftware_for_visualizing_molecular_dynamics_trajectories_image_1_v4s_c792935a28.png)\n\n\n\nAn MD run is a movie of atoms, often thousands of frames long, and one saved snapshot throws away the part that cost the compute. Three jobs sit behind any useful look at a trajectory.\n\n**Playback.** Load the run, step through frames, loop a stretch, watch a loop swing open or a ligand drift out of the pocket. That's the floor.\n\n**Quantitative analysis.** Motion says something happened; numbers say whether it matters. RMSD against a reference frame measures how far the structure wandered. A pairwise distance tracks whether a hydrogen bond held. Clustering picks out the handful of frames that stand in for the whole run.\n\n**Per-frame inspection.** Once a number points at a moment, an RMSD spike or a distance that snaps shut, you want that exact frame in front of you in 3D.\n\nSoftware that keeps those 3 jobs in one place spares you the export-and-reimport shuffle between programs.\n\n## How the common tools compare\n\n\u003Cdiv class=\"logo-row\" style=\"display:flex;flex-wrap:wrap;align-items:center;justify-content:center;gap:1.75rem;background:#fff;border-radius:12px;padding:1.25rem 1.5rem;margin:2rem 0\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fsoftware_for_visualizing_molecular_dynamics_trajectories_image_4_4ba7a7bc2c.png\" alt='VMD' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fsoftware_for_visualizing_molecular_dynamics_trajectories_image_5_3576db7056.png\" alt='PyMOL' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fsoftware_for_visualizing_molecular_dynamics_trajectories_image_7_271d452f55.png\" alt='GROMACS' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fsoftware_for_visualizing_molecular_dynamics_trajectories_image_8_bce156b0fd.png\" alt='MDAnalysis' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fsoftware_for_visualizing_molecular_dynamics_trajectories_image_9_bf45534db5.png\" alt='Schrödinger' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003C\u002Fdiv>\n\n\n\n\u003Ctable class=\"table\">\n  \u003Cthead>\n    \u003Ctr>\u003Cth>Tool\u003C\u002Fth>\u003Cth>Strength in MD work\u003C\u002Fth>\u003Cth>How Nanome relates\u003C\u002Fth>\u003C\u002Ftr>\n  \u003C\u002Fthead>\n  \u003Ctbody>\n    \u003Ctr>\n      \u003Ctd>\u003Ca href=\"https:\u002F\u002Fwww.ks.uiuc.edu\u002FResearch\u002Fvmd\u002F\">VMD\u003C\u002Fa>\u003C\u002Ftd>\n      \u003Ctd>Reads nearly every trajectory format, scripted analysis in Tcl and Python (RMSD, RDF, whatever a project needs), stays quick on very large systems\u003C\u002Ftd>\n      \u003Ctd>Nanome opens the same simulation output and plays it back for several people at once, with MARA taking the analysis request in ordinary words rather than a script\u003C\u002Ftd>\n    \u003C\u002Ftr>\n    \u003Ctr>\n      \u003Ctd>\u003Ca href=\"https:\u002F\u002Fwww.gromacs.org\">GROMACS\u003C\u002Fa>\u003C\u002Ftd>\n      \u003Ctd>Runs the simulation itself, with a deep set of command-line analysis utilities alongside it\u003C\u002Ftd>\n      \u003Ctd>Nanome sits downstream. The \u003Ccode>.gro\u003C\u002Fcode> file it writes loads standalone, and \u003Ccode>.xtc\u003C\u002Fcode> or \u003Ccode>.trr\u003C\u002Fcode> frames attach to a model already open, atom counts matching\u003C\u002Ftd>\n    \u003C\u002Ftr>\n    \u003Ctr>\n      \u003Ctd>\u003Ca href=\"https:\u002F\u002Fwww.mdanalysis.org\">MDAnalysis\u003C\u002Fa>\u003C\u002Ftd>\n      \u003Ctd>Python analysis across many trajectories at once, headless, scriptable, no GUI in the way\u003C\u002Ftd>\n      \u003Ctd>Nanome is where a result from that pipeline gets put in front of the whole project team in 3D\u003C\u002Ftd>\n    \u003C\u002Ftr>\n    \u003Ctr>\n      \u003Ctd>\u003Ca href=\"https:\u002F\u002Fpymol.org\">PyMOL\u003C\u002Fa>\u003C\u002Ftd>\n      \u003Ctd>Rendering and figure-making, multi-state files, a deep Python API\u003C\u002Ftd>\n      \u003Ctd>Nanome imports a PyMOL \u003Ccode>.pse\u003C\u002Fcode> session for viewing (import only, and QM\u002FMM link atoms can trip the load), then carries those structures into a shared session where MARA handles the trajectory math\u003C\u002Ftd>\n    \u003C\u002Ftr>\n    \u003Ctr>\n      \u003Ctd>\u003Ca href=\"https:\u002F\u002Fwww.rbvi.ucsf.edu\u002Fchimerax\u002F\">ChimeraX\u003C\u002Fa>\u003C\u002Ftd>\n      \u003Ctd>Modern rendering, morphing between states, solid trajectory support\u003C\u002Ftd>\n      \u003Ctd>Nanome reads the same everyday structure files (PDB, mmCIF, SDF, MOL2, MOL, XYZ and PQR) and adds shared XR sessions plus a copilot that names the tool behind every number\u003C\u002Ftd>\n    \u003C\u002Ftr>\n    \u003Ctr>\n      \u003Ctd>\u003Ca href=\"https:\u002F\u002Fwww.schrodinger.com\">Schrödinger\u003C\u002Fa>\u003C\u002Ftd>\n      \u003Ctd>Desmond simulations and Maestro preparation inside one commercial suite\u003C\u002Ftd>\n      \u003Ctd>Nanome imports Maestro \u003Ccode>.mae\u003C\u002Fcode> and \u003Ccode>.maegz\u003C\u002Fcode> for viewing, which is also the handoff format from \u003Ca href=\"https:\u002F\u002Fnanome.ai\u002Fintegrations\">LiveDesign\u003C\u002Fa>\u003C\u002Ftd>\n    \u003C\u002Ftr>\n    \u003Ctr>\n      \u003Ctd>Nanome, with MARA\u003C\u002Ftd>\n      \u003Ctd>Frame-by-frame playback in the browser app and in XR, MARA trajectory analysis, \u003Ca href=\"https:\u002F\u002Fnanome.ai\u002Fintegrations\">300+ integrated tools\u003C\u002Fa>, deployable inside your own network\u003C\u002Ftd>\n      \u003Ctd>The immersive, collaborative, AI-assisted layer over the MD workflow you already run\u003C\u002Ftd>\n    \u003C\u002Ftr>\n  \u003C\u002Ftbody>\n\u003C\u002Ftable>\n\n## Where Nanome fits\n\n![Two colleagues wearing ultra-thin VR headsets examine the same solid protein surface model with a visible binding pocket floating between them](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fsoftware_for_visualizing_molecular_dynamics_trajectories_image_2_v4_0870596ee4.png)\n\n\n\nTrajectory playback is a first-class part of the workspace. You bring a run in and step the frames [in a browser tab, on a Windows desktop, or in a headset (Apple Vision Pro, Meta Quest, Pico Neo, HTC Vive Focus 3)](https:\u002F\u002Fnanome.ai\u002Fsetup). Sessions hold more than one scientist, so 2 people in different time zones watch the same loop flex and point at the same residue in shared 3D.\n\nOf the MD tools on this page, Nanome is the only one where the trajectory plays inside a live multi-user session that a colleague joins from a standalone headset or a browser tab, with nothing to install.\n\nThe measuring goes through MARA. Ask for RMSD across the run and it returns the deviation per frame. Ask how far apart 2 atoms sit over time and it tracks the distance. Ask for a representative frame and it pulls the energy minima that summarize the run. MARA reports the tool behind each answer, the inputs it ran on, and the output it returned, so a second person can retrace the work rather than take a number on faith.\n\nMARA's built-in library spans 26 categories, so trajectory analysis sits beside [docking, electrostatics, ADMET prediction and structure prediction](https:\u002F\u002Fnanome.ai\u002Fagents) in the same place.\n\nOne limit worth knowing up front: surfaces are disabled during trajectory playback, because recomputing a molecular surface every frame is still too slow to stay smooth. Ribbon, stick and space-filling representations animate normally, and the surface comes back when you stop on a frame.\n\nMotion is the hardest result to put in a document, because a still frame is the one thing a trajectory isn't. A [modern molecular presentation](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fwhat-a-modern-molecular-presentation-looks-like) leans on key views pulled from the run itself.\n\nResonac spent close to 2 years inside that gap. Its computational group ran GROMACS on a vitamin C derivative alongside a set of candidate stabilizers, and the trajectories sorted the additives by shape: lauryl alcohol closed around the target as a micelle with water shut out of the interior, behenyl alcohol stacked into flat lamellar layers instead. Flattened into plots and cross-sections, those two outcomes look much alike, and the experimental group reasonably stayed with its established protocol. The two groups then met inside the same trajectories in Nanome, turned the aggregates over in 3D together, and settled the question that afternoon. An experimental iteration cycle that had taken 6 months came down to 2 or 3 days.\n\n## When a desktop tool is the better fit\n\n![A researcher examines a space-filling protein structure on a large monitor in a quiet research office.](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fsoftware_for_visualizing_molecular_dynamics_trajectories_image_3_6b98c47149.png)\n\n\n\nBatch-processing hundreds of trajectories on a headless cluster is VMD and [MDAnalysis](https:\u002F\u002Fwww.mdanalysis.org) territory. A scripted pipeline with nobody watching has no use for an interactive 3D viewer, and both handle far larger systems than a headset will render at frame rate.\n\nFor a publication-quality still or a rendered movie, PyMOL and ChimeraX have decades of rendering polish behind them and remain the safe choice.\n\nNanome is the pick when a trajectory needs more than one pair of eyes on it, in 3D, with the RMSD and distance math coming back while everyone is still looking. It plugs into [CDD Vault](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fcollaborative-drug-discovery-and-nanome-partnership-announcement), Cresset Flare and [Schrödinger](https:\u002F\u002Fwww.schrodinger.com) LiveDesign rather than standing in for a simulation engine. A structure prepared in Maestro comes over as `.mae` or `.maegz`, the same file [LiveDesign](https:\u002F\u002Fnanome.ai\u002Fintegrations) hands off, so it lands in the shared session without a conversion step. [More on supported formats](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n\nTrajectory work is one slice of a bigger picture. The broader [roundup of molecular visualization tools](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fthe-best-molecular-visualization-tools) covers the general-purpose viewers, [drug discovery software for computational chemistry](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fdrug-discovery-software-for-computational-chemistry) covers the stack around the simulation, and [software for computational antibody design](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fsoftware-for-computational-antibody-design) covers the antibody side, where a CDR loop's motion often decides the next round. Resonac's write-up and the rest live in the [case studies](https:\u002F\u002Fnanome.ai\u002Fcase-studies).\n\n## FAQ\n\n**What's the best software for molecular dynamics trajectory visualization?**\nIt depends on the job. VMD is the standard for scripted MD analysis and very large systems, while PyMOL and ChimeraX render the finest stills and movies. Nanome is the strong pick when a run has to be watched by several people at once, in immersive 3D, with RMSD, distances and representative frames coming back from a request written in plain English.\n\n**How do I visualize molecular dynamics?**\nLoad the trajectory into a viewer that plays frames, step through it, then put numbers under what you saw. In Nanome that means opening the run in the browser app or a headset, scrubbing the frames, and asking MARA for RMSD or a pairwise distance to find the moments worth a closer look.\n\n**Does Nanome do RMSD and per-frame analysis?**\nYes. MARA computes RMSD across a trajectory, tracks pairwise distances over time, and pulls representative energy-minimum frames. Every result comes back labeled with the tool that produced it and the settings it ran under.\n\n**Can I visualize trajectories without a VR headset?**\nYes. Nanome's browser app runs in a tab with nothing installed, and there's a Windows desktop build as well. Whoever is in the browser stands in the same session as colleagues wearing headsets.\n\n**Which file formats does Nanome read for MD work?**\nTrajectories use the frame-trajectory path: `.gro` loads standalone, and `.xtc`, `.trr` or `.dcd` attach to a model already open, matching its atom count, up to a 2000-frame ceiling per model. 64-bit CHARMM DCD and fixed-atom DCD fall outside that. Structure files: PDB, mmCIF and SDF; MOL, MOL2, XYZ and PQR; SMILES typed or loaded; AutoDock `.pdbqt`, which lands as PDB with charges dropped. Three session formats import for viewing only, Maestro `.mae` and `.maegz`, CCG `.moe`, and PyMOL `.pse`. A `.dx` electrostatic map overlays a loaded model rather than standing alone. Export covers PDB, SDF and SMILES only, one frame per file, so mmCIF, MAE, MOE and PSE stay read-only and a trajectory can't be written back out. There's no support for electron density maps (CCP4, MRC, DSN6) or for native LAMMPS output, where XYZ is the only bridge. [Full list](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n","2026-07-15T01:23:50.802Z","2026-09-15T16:00:09.427Z","2026-09-15T16:00:09.382Z","The best software for molecular dynamics trajectory visualization covers playback, RMSD, and per-frame analysis. Nanome does all 3 with MARA.","molecular dynamics trajectory visualization, how to visualize molecular dynamics, MD trajectory viewer, RMSD analysis, VMD, PyMOL, trajectory playback, Nanome, MARA","software-for-visualizing-molecular-dynamics-trajectories",{"id":143,"attributes":144},143,{"title":145,"content":146,"createdAt":147,"updatedAt":148,"publishedAt":149,"date":127,"description":150,"keywords":151,"slug":152,"category":17},"How to share a 3D molecular structure with your team","Sharing a 3D molecular structure with a colleague takes three things: a workspace they can open by link, a saved view for them to land on, and the link itself. The structure arrives already turned to the pocket, carrying the representations, coloring and labels you set, and the colleague can pick it up and look at it from their own angle.\n\nNanome is a collaborative molecular visualization and drug discovery platform built around that. Its saved views are called scenes. They shipped in version 2.0, gained a per-scene point of view in 2.1.1, and live in a workspace that opens in [a browser web app](https:\u002F\u002Fnanome.ai\u002Fsetup) with no install and no headset. [MARA](https:\u002F\u002Fnanome.ai\u002Fmara), the AI copilot inside Nanome, builds and arranges scenes from plain-English requests.\n\n## The 3-minute version\n\n![A researcher wearing an ultra-thin VR headset studies a ribbon-cartoon protein structure held close to the chest](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_to_share_a_3d_molecular_structure_with_your_team_image_1_v4s_046f9ad5c7.png)\n\n\n\n1. **A structure lands in a workspace.** You drop a file in, or pull an entry down from RCSB PDB by its 4-character code.\n2. **You set the view.** Representation and coloring, labels on the residues under discussion, and everything else hidden.\n3. **The view saves as a scene.** Each scene keeps its own point of view, so one can hold the whole complex while the next sits inside the pocket. Dragging reorders them. The [scenes panel docs](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_v2\u002Fscenespanel) cover the controls.\n4. **The workspace gets shared.** You pick who can open it and what they're allowed to change.\n5. **The link goes out.** Your colleague opens it in a browser, lands on the first scene at the angle you saved, and walks the rest at their own pace.\n\n## What travels, and what doesn't\n\nA coordinate file carries atoms. A shared workspace carries atoms plus the decisions made about them, which is the part a paragraph of setup notes usually has to describe.\n\nAlong with the link:\n\n- Every structure loaded in the workspace, atoms and bonds intact.\n- Representation and coloring, per entry. Cartoon, surface, sticks, ball-and-stick.\n- Labels.\n- The point of view stored on each scene, so the camera opens where it was left.\n- The order the scenes are in.\n\nThree honest limits sit on the other side of that.\n\nA workspace has no file form. It lives in the database, and there's no portable workspace file to attach to an email, so access travels as a link plus a permission and the recipient needs an account on the same deployment.\n\nWhat leaves as a file is coordinates. Export is PDB, SDF or SMILES, one frame at a time, and the camera, the coloring and the labels stay behind in the workspace.\n\nThe vendor and session formats read one way. PyMOL `.pse`, Maestro `.mae` and `.maegz`, and MOE `.moe` all open in Nanome, and none of them come back out.\n\n## How teams do this today\n\n\u003Cdiv class=\"logo-row\" style=\"display:flex;flex-wrap:wrap;align-items:center;justify-content:center;gap:1.75rem;background:#fff;border-radius:12px;padding:1.25rem 1.5rem;margin:2rem 0\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_to_share_a_3d_molecular_structure_with_your_team_image_4_693b8bdcb1.png\" alt='PyMOL' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_to_share_a_3d_molecular_structure_with_your_team_image_5_af83be40e1.png\" alt='Mol*' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_to_share_a_3d_molecular_structure_with_your_team_image_6_12e4955195.png\" alt='UCSF ChimeraX' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_to_share_a_3d_molecular_structure_with_your_team_image_7_bdbe5fa818.png\" alt='Maestro' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_to_share_a_3d_molecular_structure_with_your_team_image_8_93d77f2007.png\" alt='MOE' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003C\u002Fdiv>\n\n\n\n\u003Ctable class=\"table\">\n  \u003Cthead>\n    \u003Ctr>\u003Cth>How it goes out\u003C\u002Fth>\u003Cth>What lands with the recipient\u003C\u002Fth>\u003Cth>What they set up themselves\u003C\u002Fth>\u003C\u002Ftr>\n  \u003C\u002Fthead>\n  \u003Ctbody>\n    \u003Ctr>\u003Ctd>A PDB file by email, plus setup notes\u003C\u002Ftd>\u003Ctd>Coordinates, and a written description of the view\u003C\u002Ftd>\u003Ctd>Opening a viewer, choosing representations, finding the chain and the pocket, and working the camera toward what the notes describe\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>A screenshot in Slack\u003C\u002Ftd>\u003Ctd>One image, in seconds, readable on any device\u003C\u002Ftd>\u003Ctd>Nothing to set up. A different angle is a new request.\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>A screen-share on a call\u003C\u002Ftd>\u003Ctd>Your live view, for the length of the call\u003C\u002Ftd>\u003Ctd>Nothing during the call, and whatever they wrote down afterward\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>A PyMOL \u003Ccode>.pse\u003C\u002Fcode> session\u003C\u002Ftd>\u003Ctd>Coordinates, representations, colors and the saved scene, ready to edit\u003C\u002Ftd>\u003Ctd>Owning PyMOL, and opening the file\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>A \u003Ca href=\"https:\u002F\u002Fmolstar.org\">Mol*\u003C\u002Fa> or RCSB PDB link\u003C\u002Ftd>\u003Ctd>A public structure rendered in the browser, nothing installed\u003C\u002Ftd>\u003Ctd>The representation, wherever the link doesn't already encode a state\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>A Nanome workspace link\u003C\u002Ftd>\u003Ctd>The structures, representations, labels, and each scene's point of view\u003C\u002Ftd>\u003Ctd>Nothing beyond opening the link, with an account on the same deployment\u003C\u002Ftd>\u003C\u002Ftr>\n  \u003C\u002Ftbody>\n\u003C\u002Ftable>\n\nEach row answers a different question. A screenshot wins on speed and reaches a phone. A `.pse` is the richest thing you can hand a PyMOL user. For the full run of tools that open a coordinate file in 3D, we went through them in [how to view PDB files in 3D](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fhow-to-view-pdb-files-in-3d-and-the-tools-that-do-it-well).\n\n## Where Nanome fits\n\n![Two colleagues wearing ultra-thin VR headsets examine the same solid protein surface model with a visible binding pocket floating between them](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_to_share_a_3d_molecular_structure_with_your_team_image_2_v4s_249ce06704.png)\n\n\n\nThe unit Nanome shares is a workspace, and the thing inside it that carries an argument is the scene. A scene holds the structure, how it's drawn, what's labeled, and where the camera sits. A sequence of them carries a walkthrough: the complex, then the pocket, then the substituent everyone is arguing about, in the order the reasoning goes. [What a modern molecular presentation looks like](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fwhat-a-modern-molecular-presentation-looks-like) works through building one.\n\nBecause the workspace opens in a browser, the person on the other end needs a laptop and a link. (When people do open it at the same time, Spotlight Mode makes one person's view the shared one for everybody following, and [collaborative drug discovery software for remote teams](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fcollaborative-drug-discovery-software-for-remote-teams) covers the live side of that.)\n\nThe remaining question is usually how long it takes a first-time recipient to get their bearings once they're inside. [Kingsley et al. 2019](https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.jmgm.2019.03.010), in the Journal of Molecular Graphics and Modelling, recorded an answer while running a real project with a Novartis GNF group: new users were comfortable in the environment within a few minutes, and were pointing out candidate sites for macrocyclization shortly after.\n\nWhatever produced the structure can hand it over as it is. Nanome reads PDB, mmCIF, SDF, SMILES, MOL and MOL2, XYZ, PQR and PDBQT, along with the PyMOL, Maestro and MOE session files above, so a share doesn't open with a conversion step. [More on supported formats](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats). Teams working this way have written up what came of it at [nanome.ai\u002Fcase-studies](https:\u002F\u002Fnanome.ai\u002Fcase-studies).\n\n## When sending a file is the right pick\n\n![A flat diagram showing three sharing paths, a file, a link, and a video, each routed to a different kind of recipient.](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_to_share_a_3d_molecular_structure_with_your_team_image_3_97729d3028.png)\n\n\n\nPlenty of shares are files, and correctly so.\n\nA collaborator who works in PyMOL and wants a session to keep editing is better served by a `.pse` out of PyMOL. Nanome reads those and doesn't write them, so the round trip stops there.\n\nA deposition is a coordinate file by definition, and so is the structure that ships with a manuscript. A saved camera angle has no business in either.\n\nFor a public structure going to someone outside the org with no account, a [Mol*](https:\u002F\u002Fmolstar.org) embedded viewer or an RCSB PDB entry page beats a workspace link. The recipient needs a browser and nothing else, and there's nobody to invite. That advantage is real and it holds for anything already in the PDB.\n\nAnd if what you want to send is a rendered animation that plays on any machine, [ChimeraX](https:\u002F\u002Fwww.rbvi.ucsf.edu\u002Fchimerax\u002F) has a `movie` command that produces one, and the file it writes travels anywhere a video does.\n\n## FAQ\n\n**How do I share a protein structure with someone who doesn't have a headset?**\nA workspace link covers it. Nanome's web app runs in a browser on an ordinary laptop, so your colleague opens the link, lands on the scene you saved, and turns the structure with a mouse. Headsets are one way in among several, alongside the Windows desktop app.\n\n**Can I control which view they land on?**\nYes. Every scene stores its own point of view, which shipped in version 2.1.1, so the camera position and orientation you saved is what opens for them. Scenes are ordered and reorder by dragging, so the first thing on screen is the first thing you want talked about.\n\n**What happens to my representations and labels when I share?**\nThey travel with the scene. Cartoon, surface, sticks, coloring and labels are part of what a scene stores, along with which entries are visible and which are hidden. A file export behaves differently: a PDB, SDF or SMILES leaving Nanome carries coordinates for a single frame, and the view stays in the workspace.\n\n**Can we do this with proprietary structures?**\nYes. Enterprise deployments of Nanome, MARA included, run on [infrastructure you control](https:\u002F\u002Fnanome.ai\u002Fmara-enterprise), either as a dedicated instance or on your own hardware, so a shared workspace never leaves your environment. Projects and permissions arrived with app.nanome.ai in version 2.5, and they govern who can open a workspace and what they can change once they're in it.\n","2026-08-27T18:19:49.794Z","2026-09-15T16:00:05.926Z","2026-09-15T16:00:05.851Z","Sharing a 3D molecular structure with a colleague: a workspace link carries the saved view, representations and labels, so nothing gets rebuilt.","share a 3D molecular structure, send a protein structure, share PDB file, molecular workspace link, share structure with colleague","how-to-share-a-3d-molecular-structure-with-your-team",{"id":154,"attributes":155},144,{"title":156,"content":157,"createdAt":158,"updatedAt":159,"publishedAt":160,"date":127,"description":161,"keywords":162,"slug":163,"category":17},"How to present computational results to an experimental team","Communicating a computational result to an experimental team comes down to handing over the 3D object the calculation was about, along with the evidence behind it, so the people who run the next experiment can turn it and ask their own questions. Nanome is a collaborative molecular visualization and drug discovery platform that covers that handoff. It plays back molecular dynamics trajectories frame by frame, saves 3D views of a structure as scenes, and hosts shared sessions colleagues join from [a browser tab or a headset](https:\u002F\u002Fnanome.ai\u002Fsetup). [MARA](https:\u002F\u002Fnanome.ai\u002Fmara), the AI copilot inside it, runs the analysis when a question lands mid-discussion, so the number and the molecule arrive together.\n\nMost results travel as a plot, a slide and a paragraph of interpretation, and for plenty of results that's the right form. The ones that suffer are the results whose content is a shape.\n\n## Where a computational result loses people\n\n![Two colleagues wearing ultra-thin VR headsets examine the same ribbon-cartoon protein structure floating between them](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fpresenting_computational_results_to_experimental_teams_image_1_v4s_9cee1d2184.png)\n\n\n\nA simulation is a 3D object moving through time. A figure of it is a photograph, taken from one camera angle, picked before anyone in the room had asked a question.\n\nFor a quantity, that works out fine. For a geometry, it gets expensive. Somebody who wants to know what the pocket looks like from behind the ligand, or whether 2 aggregates differ in shape or only in size, is asking for a different photograph, and producing one means going back to the workstation and finding another hour on 2 calendars.\n\nThe people on either side of that handoff have spent their hours differently. Whoever ran the simulation has turned the system over hundreds of times and knows what each projection drops. Whoever runs the next experiment is meeting the object for the first time in that one figure, and the detail a projection drops is often the detail the experiment turns on.\n\n## What survives the trip, and what flattens\n\nSome results lose almost nothing on the way to a flat page. Others lose the part that made them worth the compute.\n\n\u003Ctable class=\"table\">\n  \u003Cthead>\n    \u003Ctr>\u003Cth>Result\u003C\u002Fth>\u003Cth>What the 2D version carries\u003C\u002Fth>\u003Cth>What the 3D version adds\u003C\u002Fth>\u003C\u002Ftr>\n  \u003C\u002Fthead>\n  \u003Ctbody>\n    \u003Ctr>\u003Ctd>A docked pose\u003C\u002Ftd>\u003Ctd>An interaction diagram with contacts drawn as dashes, plus a render from one camera angle\u003C\u002Ftd>\u003Ctd>The pocket at true depth, where a clash reads as a clash and an H-bond angle can be checked from either side\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>An MD trajectory\u003C\u002Ftd>\u003Ctd>RMSD or a distance plotted against time, with a handful of representative snapshots\u003C\u002Ftd>\u003Ctd>The motion itself, frame by frame, with the loop or the ligand moving in front of the group\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>An aggregation or packing difference\u003C\u002Ftd>\u003Ctd>A cross-section and a radial distribution function, both framed before the question came up\u003C\u002Ftd>\u003Ctd>The assembly turned over in the hand, where a closed cluster and a stacked layer separate on sight\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>A chain interface\u003C\u002Ftd>\u003Ctd>Buried surface area as a number, plus a contact map\u003C\u002Ftd>\u003Ctd>The shape of the interface, and which residues actually sit across from which\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>A conformational change\u003C\u002Ftd>\u003Ctd>2 overlaid ribbons, or a morph rendered along one camera path\u003C\u002Ftd>\u003Ctd>Both states held side by side and rotated together, at the scale the change happens on\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>RMSD over the course of a run\u003C\u002Ftd>\u003Ctd>A line, which is what the quantity is\u003C\u002Ftd>\u003Ctd>Very little. The 3D view comes in afterwards, once a spike names a frame worth opening\u003C\u002Ftd>\u003C\u002Ftr>\n  \u003C\u002Ftbody>\n\u003C\u002Ftable>\n\nThat last row belongs there as much as the other 5. RMSD against a reference frame is a scalar moving along one axis, and a line is the right way to draw it.\n\n## How teams do it\n\n\u003Cdiv class=\"logo-row\" style=\"display:flex;flex-wrap:wrap;align-items:center;justify-content:center;gap:1.75rem;background:#fff;border-radius:12px;padding:1.25rem 1.5rem;margin:2rem 0\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fpresenting_computational_results_to_experimental_teams_image_4_43b661e96e.png\" alt='GROMACS' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fpresenting_computational_results_to_experimental_teams_image_5_bf01284ad1.png\" alt='MDAnalysis' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fpresenting_computational_results_to_experimental_teams_image_6_bf8f3538ee.png\" alt='VMD' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fpresenting_computational_results_to_experimental_teams_image_7_4926c731e2.png\" alt='PyMOL' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fpresenting_computational_results_to_experimental_teams_image_8_634e2a6a21.png\" alt='UCSF ChimeraX' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fpresenting_computational_results_to_experimental_teams_image_9_66da1ae2a6.png\" alt='Mol*' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003C\u002Fdiv>\n\n\n\n\u003Ctable class=\"table\">\n  \u003Cthead>\n    \u003Ctr>\u003Cth>How the result travels\u003C\u002Fth>\u003Cth>Strongest at\u003C\u002Fth>\u003Cth>What the audience can do with it\u003C\u002Fth>\u003C\u002Ftr>\n  \u003C\u002Fthead>\n  \u003Ctbody>\n    \u003Ctr>\u003Ctd>\u003Ca href=\"https:\u002F\u002Fwww.gromacs.org\">GROMACS\u003C\u002Fa> or \u003Ca href=\"https:\u002F\u002Fwww.mdanalysis.org\">MDAnalysis\u003C\u002Fa> analysis plus a plot\u003C\u002Ftd>\u003Ctd>Numbers that are reproducible, scriptable and ready for a manuscript\u003C\u002Ftd>\u003Ctd>Reading the value and checking the axes. A question about a different frame goes back into the run queue\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>A \u003Ca href=\"https:\u002F\u002Fwww.ks.uiuc.edu\u002FResearch\u002Fvmd\u002F\">VMD\u003C\u002Fa> movie\u003C\u002Ftd>\u003Ctd>Smooth playback of very large systems, scripted and repeatable\u003C\u002Ftd>\u003Ctd>Watching the camera path the author picked, at the pace the author picked\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>A \u003Ca href=\"https:\u002F\u002Fpymol.org\">PyMOL\u003C\u002Fa> session\u003C\u002Ftd>\u003Ctd>Representations, selections and saved views preserved in one file\u003C\u002Ftd>\u003Ctd>Opening it and turning the structure, for anyone who owns PyMOL\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>A slide deck\u003C\u002Ftd>\u003Ctd>Travels anywhere, archives cleanly, opens on any laptop\u003C\u002Ftd>\u003Ctd>Following the argument. Each structure stays the photograph it was exported as\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Nanome scenes in a shared session\u003C\u002Ftd>\u003Ctd>Saved 3D views of the live structure, walked through together on the web or in a headset\u003C\u002Ftd>\u003Ctd>Turning the molecule from their own angle, and getting a number back from MARA without leaving the view\u003C\u002Ftd>\u003C\u002Ftr>\n  \u003C\u002Ftbody>\n\u003C\u002Ftable>\n\nThe first 4 rows all end in one place: the audience receives a rendering of a decision that was made in 3D somewhere else.\n\n## Where Nanome fits\n\n![Two colleagues wearing ultra-thin VR headsets examine the same space-filling protein model floating between them](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fpresenting_computational_results_to_experimental_teams_image_2_v4_c3fd255f2d.png)\n\n\n\nTrajectory playback shipped in Nanome 2.5. A run opens in the workspace and steps frame by frame in a browser tab, on Windows, or in a headset, and a session holds more than one person, so the computational scientist and the bench scientist watch the same frames at the same moment from wherever they each are. [Trajectory playback docs](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Fmdplayback).\n\nScenes give the argument a running order. A scene is a saved 3D view of the real structure carrying its own point of view, representations and labels, and scenes sit in a sequence the way slides do, reorderable by dragging. [Scenes docs](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_v2\u002Fscenespanel). Everyone opens on the same view, and from there each person can turn the molecule and study the part they came for. That is the request a flat figure has to send back to the workstation.\n\nQuestions that arrive mid-discussion go to MARA. RMSD across the run, the distance between 2 atoms over time, a representative frame from the energy minima: the answer lands on the structure the group is already looking at, labeled with the tool that produced it and the inputs it ran on, so whoever asked can retrace it later.\n\nOne limit belongs in the open. Surfaces are disabled during trajectory playback, because recomputing a molecular surface at every frame costs too much to hold a steady frame rate today. A run plays back as ribbon, stick or space-filling, and the surface returns the moment playback stops.\n\nResonac's computational and experimental groups worked in parallel for close to 2 years while the molecular dynamics evidence lived in trajectories and 2D plots. The simulations covered roughly 200 APPS molecules in 150,000 waters: lauryl alcohol packed into micelle-like clusters, while behenyl alcohol drove layered lamellar structures. On a flat plot those 2 arrangements are hard to tell apart. The 2 groups then stepped through the same GROMACS trajectories together in 3D, and aligned within a few hours.\n\nSeparately, an experimental iteration cycle at Resonac that had taken 6 months came down to roughly 2 to 3 days. The write-up sits with the others at [nanome.ai\u002Fcase-studies](https:\u002F\u002Fnanome.ai\u002Fcase-studies).\n\nRelated questions have their own pages. [Software for visualizing molecular dynamics trajectories](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fsoftware-for-visualizing-molecular-dynamics-trajectories) compares the viewers side by side. How a set of scenes gets built and handed on is the subject of [what a modern molecular presentation looks like](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fwhat-a-modern-molecular-presentation-looks-like). When the 2 groups sit in different time zones, [collaborative drug discovery software for remote teams](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fcollaborative-drug-discovery-software-for-remote-teams) covers how a live session runs across sites.\n\n## When a plot is the right answer\n\nA lot of computational output is one-dimensional, and drawing it in 3D adds nothing.\n\nRMSD over time is a line. So is a distance between 2 atoms, a free energy along a reaction coordinate, a potency series, an ADMET table. When the content of a result is a number moving along one axis, a plot states it exactly.\n\nA manuscript figure, a regulatory submission and a summary slide in a board deck all want one fixed image that will look the same in 10 years. For a rendered movie with a scripted camera, [ChimeraX](https:\u002F\u002Fwww.rbvi.ucsf.edu\u002Fchimerax\u002F) has a genuine `movie` command and decades of rendering work behind it, while VMD and PyMOL still turn out the stills that hold up in print.\n\nFor a public structure that simply has to reach somebody, a [Mol\\*](https:\u002F\u002Fmolstar.org) viewer embedded in a page or a wiki wins outright. The recipient clicks a link, the structure loads in their browser, and no account and no software sit in the way. RCSB PDB serves Mol\\* on every entry page, which covers a large share of \"can you show me this protein\" traffic, at zero cost.\n\nNanome is the pick when the content of the result is a shape or a motion, when the people who have to act on it need to look from their own angle, and when the follow-up question deserves an answer while everyone is still in the room.\n\n## FAQ\n\n**How do you explain molecular dynamics results to someone who doesn't run simulations?**\nLeading with the physical picture and keeping the numbers beside it tends to travel furthest. The motion is the finding, so a trajectory that actually plays, with the loop or the ligand moving, carries an audience further than a plot introduced first. In Nanome that means opening the run in a shared session, stepping the frames together, and asking MARA for the RMSD or the distance that puts a number under what everyone just watched.\n\n**What trajectory formats can be reviewed in 3D?**\nA GROMACS `.gro` file loads on its own. Frames in `.xtc`, `.trr` and `.dcd` attach to a model that is already open, and their atom count has to equal that model's. All 4 run on the frame-trajectory path, which tops out at a 2000-frame ceiling per model, and 64-bit CHARMM DCD and fixed-atom DCD fall outside it. [More on supported formats](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n\n**Does everyone need a headset to review a trajectory together?**\nNo. The browser web app runs in a tab with nothing to install, and there's a Windows desktop build as well. Someone in a browser stands in the same session as a colleague in a Meta Quest, Apple Vision Pro, Pico Neo or HTC Vive Focus 3, watching the same frames. Headsets carry depth and scale a monitor can't, and the browser puts everybody else in the discussion.\n\n**How long does it take to set up a review like this?**\nThe prep has the same shape as building slides. The trajectory or the structure loads into the workspace, a scene gets saved at each view the discussion needs, and the scenes get dragged into the order the argument runs in. Whoever is joining opens a link in a browser. The step that usually eats the time, converting a file so the other side can open it, drops out once both sides are in the same workspace.\n","2026-08-27T18:19:50.150Z","2026-09-15T16:00:11.153Z","2026-09-15T16:00:11.120Z","How computational chemists get a docked pose, an MD trajectory or a packing difference across to a bench team, and when a plot is the better medium.","communicate computational chemistry results, explain molecular dynamics results, computational experimental collaboration, present MD results","presenting-computational-results-to-experimental-teams",{"id":165,"attributes":166},91,{"title":167,"content":168,"createdAt":169,"updatedAt":170,"publishedAt":171,"date":172,"description":173,"keywords":174,"slug":175,"category":17},"How Nanome is different from UCSF ChimeraX","If you're looking for alternatives to ChimeraX, the short version is that Nanome and [UCSF ChimeraX](https:\u002F\u002Fwww.rbvi.ucsf.edu\u002Fchimerax\u002F) solve overlapping problems in different ways. ChimeraX is a free, powerful academic desktop program for viewing and analyzing molecular structures. Nanome is a collaborative molecular visualization and drug discovery platform that runs on a browser web app and XR headsets, and it opens the same PDB and SDF files ChimeraX does. Nanome also carries an AI copilot called [MARA](https:\u002F\u002Fnanome.ai\u002Fmara), which takes analysis requests in plain English.\n\nThe useful question is where each one fits.\n\n## What ChimeraX is good at\n\n![A researcher in a modern office studies a ribbon-rendered protein structure on a large display.](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_nanome_is_different_from_ucsf_chimerax_image_1_ddb5852c25.png)\n\n\n\nChimeraX is a mature desktop tool with deep analysis features, high-quality rendering, and a scripting command line. It's free for academic use, it's actively developed at UCSF, and a lot of structural biologists know it cold.\n\nFor a single-user program aimed at careful figure-making, density map work, or command-driven analysis, ChimeraX is a strong pick, and Nanome doesn't try to reproduce it.\n\n## Where Nanome is different\n\n![Two colleagues wearing ultra-thin VR headsets examine the same solid protein surface model with a visible binding pocket floating between them](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_nanome_is_different_from_ucsf_chimerax_image_2_v4_108a4e637e.png)\n\n\n\nNanome is built around three things.\n\n**Real-time multiplayer.** Several people occupy one molecular scene at the same time and point at the same atoms while they talk it through. Review and design happen in a shared room rather than over screen-share.\n\n**Native XR.** Nanome [renders structures in immersive 3D on Meta Quest, Apple Vision Pro, Pico Neo, and HTC Vive Focus 3, plus Windows desktop](https:\u002F\u002Fnanome.ai\u002Fsetup). Walking around a binding pocket at arm's length reads differently from spinning it with a mouse.\n\n**An AI copilot.** MARA takes a docking run or a fold prediction as a request in ordinary words, picks the engine, runs the job, and logs which one it used and on what inputs, so the work stays checkable afterwards. Its library spans [26 categories and more than 300 tools](https:\u002F\u002Fnanome.ai\u002Fintegrations): Smina and DiffDock-L for docking, AlphaFold 3, OpenFold3, and [Boltz-2](https:\u002F\u002Fgithub.com\u002Fjwohlwend\u002Fboltz) for folding and co-folding, APBS for electrostatics, ADMET and toxicity models, ProteinMPNN for sequence design, ANARCI for numbering antibody variable domains and marking their CDR loops, and RFdiffusion3 (beta) for de novo binders.\n\nThere's a browser web app as well, so a laptop is enough to join.\n\n## Side by side\n\n\u003Ctable class=\"table\">\n  \u003Cthead>\n    \u003Ctr>\u003Cth>Program\u003C\u002Fth>\u003Cth>Where it's strong\u003C\u002Fth>\u003Cth>How Nanome sits with it\u003C\u002Fth>\u003C\u002Ftr>\n  \u003C\u002Fthead>\n  \u003Ctbody>\n    \u003Ctr>\u003Ctd>UCSF ChimeraX\u003C\u002Ftd>\u003Ctd>Free for academic use: desktop viewing, analysis, rendering, and a scripting command line, plus VR and multi-person meeting sessions\u003C\u002Ftd>\u003Ctd>Nanome opens the same PDB and SDF files and puts them in a session people join from a browser tab or a standalone headset\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>UCSF Chimera (legacy)\u003C\u002Ftd>\u003Ctd>The older desktop predecessor, still in use in some labs\u003C\u002Ftd>\u003Ctd>Nanome reads the same standard structure files and fetches from RCSB PDB, PubChem, and DrugBank\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Nanome\u003C\u002Ftd>\u003Ctd>Shared sessions, native XR, and MARA driving analysis from plain English\u003C\u002Ftd>\u003Ctd>A layer for group review and design on top of whatever writes the structures\u003C\u002Ftd>\u003C\u002Ftr>\n  \u003C\u002Ftbody>\n\u003C\u002Ftable>\n\n## What ChimeraX already does here\n\nChimeraX covers more of this ground than most desktop tools, and it's worth being\nspecific about how much. `coordset` plays a trajectory. The `vr` command renders the\nscene in a headset. And `meeting` shares one scene between ChimeraX instances running on\ndifferent computers, including in VR, so two people in different buildings can look at\nthe same structure at the same time. That's a real multi-user feature, built by a group\nthat has been doing this longer than we have.\n\nThe difference shows up in what each participant has to bring. A ChimeraX meeting asks\neveryone in it to install ChimeraX and connect to port 52194 on the host machine, with\nchimeraxmeeting.net available as a relay when the host sits behind a firewall. The VR\nside is a tethered setup: SteamVR or OpenXR, a headset wired to a Windows PC, and a GPU\nwith real headroom behind it. UCSF's own VR documentation is candid about the ceiling,\nnoting that structures past a few thousand atoms render slowly enough to stutter in the\nheadset. A solvated MD system is well past that.\n\nThere was a standalone path for a while. UCSF's LookSee app sent a ChimeraX scene one\nway to a Quest, under a triangle budget, as a viewer rather than a session. It has since\nbeen discontinued.\n\nNanome starts from the person joining. The trajectory plays in a live multi-user\nsession, and a colleague comes in from a standalone Quest, Pico Neo, Vive Focus 3, or\nApple Vision Pro, or from a browser tab on a laptop, with nothing installed on their\nend. Of the tools in this comparison, Nanome is the only one where that's true.\n\n## Files and fit\n\n\u003Cdiv class=\"logo-row\" style=\"display:flex;flex-wrap:wrap;align-items:center;justify-content:center;gap:1.75rem;background:#fff;border-radius:12px;padding:1.25rem 1.5rem;margin:2rem 0\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_nanome_is_different_from_ucsf_chimerax_image_5_f6a10650c8.png\" alt='Schrödinger Maestro' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_nanome_is_different_from_ucsf_chimerax_image_6_5a10495f59.png\" alt='OpenEye' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_nanome_is_different_from_ucsf_chimerax_image_7_25e46752dc.png\" alt='Cresset Flare' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_nanome_is_different_from_ucsf_chimerax_image_8_1c31c360fc.png\" alt='CDD Vault' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_nanome_is_different_from_ucsf_chimerax_image_9_7bd2a0604d.png\" alt='GROMACS' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003C\u002Fdiv>\n\n\n\n![A diagram showing many molecular file formats converging as arrows into a single platform, with PDB, SDF, and SMILES flowing back out as exports.](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_nanome_is_different_from_ucsf_chimerax_image_3_r3_27db61c896.png)\n\n\n\nStructures cross between the two without a conversion step in the middle. Nanome imports `.pdb` and `.ent`, mmCIF as `.cif`, `.mmcif`, `.mcif`, or `.bcif`, plus `.sdf`, `.mol`, `.mol2`, SMILES, `.xyz`, `.pqr`, and `.pdbqt`, and it fetches by accession from RCSB PDB, PubChem, DrugBank, ChEMBL, and UniProt. What leaves Nanome is PDB, SDF, or SMILES, a single frame at a time. [More on supported formats](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n\nNanome also sits beside commercial software a group already licenses. On the [Schrödinger](https:\u002F\u002Fwww.schrodinger.com) side, the LiveDesign gadget hands [Maestro](https:\u002F\u002Fwww.schrodinger.com\u002Fplatform\u002Fproducts\u002Fmaestro\u002F) `.mae` and `.maegz` files straight across. [OpenEye (Cadence)](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fview-your-openeyecadence-data-with-fresh-eyes-using-nanome-xr), Cresset Flare, and [CDD Vault](https:\u002F\u002Fwww.collaborativedrug.com) connect too, which is why it usually joins a stack instead of displacing one.\n\nFor molecular dynamics, Nanome plays trajectories back from [GROMACS](https:\u002F\u002Fwww.gromacs.org) and other engines. A `.gro` file loads on its own. The `.xtc`, `.trr`, and `.dcd` formats attach to a model that's already open and have to carry the same atom count, and the frame-trajectory substrate holds a 2000-frame ceiling.\n\nOne piece of this comparison has been measured. A 2019 benchmark in the Journal of Molecular Graphics and Modelling, [Kingsley et al.](https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.jmgm.2019.03.010), clocked surface rendering across Nanome and the then-current ChimeraX, Chimera, PyMOL, and Discovery Studio builds, and Nanome came out several times faster while drawing two eye views at 90+ frames per second. Novartis GNF scientists are on the author list. Every program in that test has shipped years of releases since, which makes it a 2019 measurement of 2019 builds.\n\n## When ChimeraX is the right pick\n\nChimeraX is the better fit for a lab that wants a workstation program free for academic use, for detailed density map or rendering work, or where the group already standardizes on its command line.\n\nNanome fits when the work is collaborative, when depth perception helps read a pocket or a loop, or when an analysis is easier to ask for in words than to script.\n\nThree nearby comparisons run along the same seam. [How Nanome differs from PyMOL](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fhow-nanome-is-different-from-pymol) is the hub for the set and covers the other free viewer most labs have installed. [Discovery Studio](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fhow-nanome-is-different-from-biovia-discovery-studio) covers a broad commercial suite, and [MOE](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fhow-nanome-is-different-from-moe) covers one that doubles as a Nanome integration. The [case studies](https:\u002F\u002Fnanome.ai\u002Fcase-studies) show how groups have split the work in practice.\n\n## FAQ\n\n**What are good alternatives to ChimeraX for team work?**\nNanome is one. ChimeraX has a `meeting` command that shares a scene between ChimeraX instances, including in VR, and it works well when everyone already runs ChimeraX on a capable machine. Nanome's sessions are built for the case where they don't: a colleague joins from a standalone headset or a browser tab with nothing to install, and several people share the same structure at once.\n\n**Is Nanome a ChimeraX alternative if I only have a laptop?**\nYes. The browser web app needs no headset and no install, and it opens the same PDB and SDF files, so shared sessions and MARA come without XR hardware.\n\n**Do I have to stop using ChimeraX to use Nanome?**\nNo. Both read the same standard structure files, and Nanome fetches from RCSB PDB, PubChem, DrugBank, ChEMBL, and UniProt. Plenty of groups keep a desktop viewer for solo analysis and run Nanome for shared review and design.\n\n**What about alternatives to UCSF Chimera, the older version?**\nNanome works there too. It opens the same standard files legacy Chimera handles, and adds shared sessions, XR, and MARA on top.\n\n**What file formats does Nanome support?**\nView and edit: `.pdb` and `.ent`, mmCIF (`.cif`, `.mmcif`, `.mcif`, `.bcif`), `.sdf`, `.sd`, `.mol`, `.mol2`, SMILES (typed or `.smi`), `.xyz`, `.pqr`, and `.pdbqt`. View only, from other vendors: Maestro `.mae` and `.maegz`, `.moe`, and PyMOL `.pse`. Animation runs two ways: multi-MODEL PDB, mmCIF, multi-record SDF, and multi-block XYZ step through whole models, while `.gro`, `.xtc`, `.trr`, and `.dcd` step through coordinate frames. Electrostatic maps arrive as `.dx` and overlay a model already loaded. Writing back out is PDB, SDF, or SMILES, one frame, which leaves mmCIF, MAE, MOE, and PSE import-only. [More on supported formats](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n","2026-07-15T01:23:51.028Z","2026-09-10T16:00:10.067Z","2026-09-10T16:00:10.019Z","2026-09-10","A fair look at Nanome as a ChimeraX alternative: real-time multiplayer, native XR, an AI copilot, and a web app that opens the same files.","alternatives to ChimeraX, ChimeraX alternative, alternatives to UCSF Chimera, UCSF ChimeraX, molecular visualization, Nanome, MARA","how-nanome-is-different-from-ucsf-chimerax",{"id":177,"attributes":178},98,{"title":179,"content":180,"createdAt":181,"updatedAt":182,"publishedAt":183,"date":172,"description":184,"keywords":185,"slug":186,"category":17},"Tools for visualizing protein-ligand interactions","The best protein-ligand interaction visualization tools include [PyMOL](https:\u002F\u002Fpymol.org), [BIOVIA Discovery Studio](https:\u002F\u002Fwww.3ds.com\u002Fproducts\u002Fbiovia\u002Fdiscovery-studio), and [Schrödinger Maestro](https:\u002F\u002Fwww.schrodinger.com\u002Fplatform\u002Fproducts\u002Fmaestro\u002F) on the desktop, plus Nanome for interactive 3D and XR inspection. Nanome is a collaborative molecular visualization and drug discovery platform that runs in the browser and on XR headsets, so you can walk around a binding pocket in real 3D. [MARA](https:\u002F\u002Fnanome.ai\u002Fmara), the AI copilot inside it, computes the actual interactions on request. The right pick depends on whether you want a publication figure, a full comp-chem workflow, or a hands-on look at how a ligand sits in its pocket.\n\n## What these tools show you\n\n![A researcher wearing an ultra-thin VR headset studies a solid protein surface model with a small ligand visible in its binding pocket held close to the chest](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Ftools_for_visualizing_protein_ligand_interactions_image_1_v4_f80724355e.png)\n\n\n\nA protein-ligand interaction map turns a pile of atoms into the specific contacts that make a ligand bind. The ones worth reading:\n\n- **Hydrogen bonds:** donor and acceptor pairs, usually drawn as dashed lines with distances in angstroms.\n- **Hydrophobic contacts:** where greasy parts of the ligand tuck against nonpolar residues.\n- **Pi-stacking:** aromatic rings stacking face-to-face or edge-to-face, common with tryptophan, phenylalanine, tyrosine, and histidine.\n- **Salt bridges and electrostatics:** charged groups pulling on each other.\n- **Steric clashes:** atoms sitting too close, a red flag on a docked or predicted pose.\n\nA good tool labels these, measures them, and lets you toggle each type so the picture doesn't turn to spaghetti.\n\nContact reading is one narrow job a viewer does, and the wider survey of what's out there sits in [the best molecular visualization tools](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fthe-best-molecular-visualization-tools).\n\n## Where each tool fits\n\n\u003Cdiv class=\"logo-row\" style=\"display:flex;flex-wrap:wrap;align-items:center;justify-content:center;gap:1.75rem;background:#fff;border-radius:12px;padding:1.25rem 1.5rem;margin:2rem 0\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Ftools_for_visualizing_protein_ligand_interactions_image_4_4f0e77e5c3.png\" alt='PyMOL' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Ftools_for_visualizing_protein_ligand_interactions_image_6_784c306019.png\" alt='Schrödinger Maestro' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Ftools_for_visualizing_protein_ligand_interactions_image_7_4725a916cc.png\" alt='OpenEye' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Ftools_for_visualizing_protein_ligand_interactions_image_8_17211f7dc9.png\" alt='Cresset Flare' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Ftools_for_visualizing_protein_ligand_interactions_image_9_073b4fe49a.png\" alt='CDD Vault' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003C\u002Fdiv>\n\n\n\n\u003Ctable class=\"table\">\n  \u003Cthead>\n    \u003Ctr>\u003Cth>Tool\u003C\u002Fth>\u003Cth>Its strength\u003C\u002Fth>\u003Cth>How Nanome sits next to it\u003C\u002Fth>\u003C\u002Ftr>\n  \u003C\u002Fthead>\n  \u003Ctbody>\n    \u003Ctr>\u003Ctd>PyMOL\u003C\u002Ftd>\u003Ctd>Rendered figures for manuscripts, scripted from Python, cited across the literature\u003C\u002Ftd>\u003Ctd>Nanome opens the saved \u003Ccode>.pse\u003C\u002Fcode> session so a pocket becomes something two people can walk into together\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>BIOVIA Discovery Studio\u003C\u002Ftd>\u003Ctd>Desktop suite with 2D interaction diagrams and contact tables\u003C\u002Ftd>\u003Ctd>MARA redraws those same contacts in 3D on request and names the routine that produced them\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Schrödinger Maestro\u003C\u002Ftd>\u003Ctd>Docking, free energy, ligand prep, validated end to end\u003C\u002Ftd>\u003Ctd>Nanome connects to \u003Ca href=\"https:\u002F\u002Fnanome.ai\u002Fintegrations\">Schrödinger LiveDesign\u003C\u002Fa>, so a pose crosses over instead of being rebuilt\u003C\u002Ftd>\u003C\u002Ftr>\n  \u003C\u002Ftbody>\n\u003C\u002Ftable>\n\nPyMOL is where most published figures come from. It's precise, it scripts cleanly, and almost anyone in the field can open a session someone sends them. Nanome reads those `.pse` files directly, though a session carrying QM\u002FMM link atoms can trip the parser.\n\nDiscovery Studio and Maestro are heavier suites covering the whole pipeline, prep through scoring, with contact analysis built in. Structures cross into Nanome as Maestro `.mae` and `.maegz` (the pair LiveDesign hands off), and the everyday structure formats load too, PDB and mmCIF through SDF, MOL2, XYZ and PQR, with `.pdbqt` added in 2.6.0. [More on supported formats](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n\nAny of those three is the better answer when the deliverable is a particular rendering, a validated docking protocol, or a 2D contact diagram for a report. They're mature and they're accurate. Where the interaction read sits inside a wider computational stack is mapped out in [drug discovery software for computational chemistry](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fdrug-discovery-software-for-computational-chemistry).\n\n## Where Nanome is different\n\n![Two colleagues wearing ultra-thin VR headsets examine the same ribbon-cartoon protein structure floating between them](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Ftools_for_visualizing_protein_ligand_interactions_image_2_v4_aec25a3d11.png)\n\n\n\nNanome puts you inside the structure. You load a PDB or SDF file, or fetch a code from RCSB PDB, PubChem or DrugBank, and the pocket turns into something you can lean into and rotate with your hands on an [Apple Vision Pro, Meta Quest, Pico Neo, HTC Vive Focus 3, a Windows desktop, or the browser web app](https:\u002F\u002Fnanome.ai\u002Fsetup) with no headset.\n\nNimbus Therapeutics changed a decision on the strength of that view. The team was evaluating the AMPKβ2 enzyme and had settled on a selectivity strategy. Seeing the protein flex in VR surfaced a better synthetic vector, and they redirected the chemistry. Their summary of the outcome: \"We were able to make our compounds more active on the target.\"\n\nMARA is what changes the mechanics of inspecting a contact. Ask in plain English for a [report of the contacts between a ligand and its protein](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_v2\u002Ftoolspanel), and it picks the routine, draws the hydrogen bonds, hydrophobic contacts and pi-stacking, and records which routine it called and on what. The same session goes further when the question does: docking, electrostatics through APBS, ADMET prediction, or a co-fold on [Boltz-2](https:\u002F\u002Fgithub.com\u002Fjwohlwend\u002Fboltz), one of several engines it can call for that job. MARA's built-in library runs to [300+ tools across 26 categories](https:\u002F\u002Fnanome.ai\u002Fintegrations).\n\nThe same reading applies at an antibody-antigen interface, where the contacts sit between a CDR loop and its epitope. That lane gets its own treatment in [software for computational antibody design](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fsoftware-for-computational-antibody-design).\n\nTwo people can stand in the same pocket at once. Real-time multiplayer puts a chemist and a structural biologist on the same clash, on the same atoms, working it out while both are still looking at it.\n\n## When another tool is the right call\n\n![A flat vector diagram showing four tool icons connected by arrows, with the collaborative inspection hub linked back to each specialized tool to illustrate integration rather than replacement.](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Ftools_for_visualizing_protein_ligand_interactions_image_3_b44321e0ff.png)\n\n\n\nFor a static, pixel-perfect figure headed into a manuscript, PyMOL remains the right answer. For work that already lives inside a validated Schrödinger or Discovery Studio pipeline, scoring and prep belong where they are. Nanome connects to Schrödinger LiveDesign, [OpenEye](https:\u002F\u002Fwww.eyesopen.com), Cresset Flare and CDD Vault, so it takes a seat beside those tools rather than asking anyone to give them up.\n\nNanome is the pick for turning a pose over in true 3D, catching a clash that a flat projection hides, or bringing a collaborator into the pocket alongside you. Projects that worked that way are written up in the [case studies](https:\u002F\u002Fnanome.ai\u002Fcase-studies).\n\n## FAQ\n\n**What is the best tool for visualizing protein-ligand interactions?**\nFor a figure in a paper, PyMOL. For a full comp-chem suite, Schrödinger Maestro or BIOVIA Discovery Studio. For interactive 3D and XR inspection with MARA computing the hydrogen bonds, hydrophobic contacts and pi-stacking, Nanome. Most groups run more than 1 of these, and Nanome connects to several.\n\n**Can Nanome show hydrogen bonds and pi-stacking?**\nYes. Ask MARA for the interactions between a ligand and its protein and it draws hydrogen bonds, hydrophobic contacts and pi-stacking, flags steric clashes, and names the routine behind the result along with what it was given.\n\n**What file formats does Nanome open?**\nFormats that import and stay editable: SDF (`.sdf`, `.sd`), PDB (`.pdb`, `.ent`), mmCIF (`.cif`, `.mmcif`, `.mcif`, `.bcif`), MOL and MOL2, XYZ, PQR, plus SMILES typed in directly. Formats that import for viewing: PDBQT, which converts to PDB with charges dropped, Maestro `.mae` and `.maegz`, MOE `.moe`, and PyMOL `.pse`. What comes back out is one frame as PDB, SDF or SMILES, so mmCIF, MAE, MOE and PSE travel one way in. The full table lives in the [supported formats reference](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n\n**Does Nanome need a VR headset?**\nNo. It runs in a browser web app with nothing to install, and there's a Windows desktop build. Meta Quest and Apple Vision Pro add the hands-on version of the same session.\n\n**Does Nanome replace PyMOL or Schrödinger?**\nNo, it works next to them. Nanome reads the files those suites already write and links up with Schrödinger LiveDesign, OpenEye, Cresset Flare and [CDD Vault](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fcollaborative-drug-discovery-and-nanome-partnership-announcement), so a pose and its data travel across rather than getting rebuilt.\n","2026-07-15T01:23:51.859Z","2026-09-10T16:00:14.400Z","2026-09-10T16:00:14.311Z","The best protein-ligand interaction visualization tools, from PyMOL to Nanome's interactive 3D and XR inspection with MARA.","protein-ligand interaction visualization tools, protein ligand interactions, hydrogen bonds, pi-stacking, PyMOL, Discovery Studio, Schrodinger, Nanome, MARA","tools-for-visualizing-protein-ligand-interactions",{"id":188,"attributes":189},96,{"title":190,"content":191,"createdAt":192,"updatedAt":193,"publishedAt":194,"date":172,"description":195,"keywords":196,"slug":197,"category":17},"How Nanome is different from VMD","Nanome is a collaborative molecular visualization and drug discovery platform that runs across a browser web app and XR headsets. It carries an [AI copilot called MARA](https:\u002F\u002Fnanome.ai\u002Fmara) that runs scientific tools from requests written in plain English. [VMD](https:\u002F\u002Fwww.ks.uiuc.edu\u002FResearch\u002Fvmd\u002F), from the University of Illinois, is free for non-commercial and internal use, and molecular dynamics groups trust it for looking at and measuring trajectories, with a scripting console deep enough to write whatever analysis a project needs. The two overlap on trajectories and separate on everything around them: who else stands in the scene, what hardware draws it, and how an analysis gets asked for. Which one fits a given week depends on whether the hard part is the analysis or getting the rest of the project to see what it showed.\n\nBoth open the files you already have. Nanome reads PDB and mmCIF, SDF and MOL2, plus XYZ and PQR, and it fetches structures straight from RCSB PDB, PubChem and DrugBank. Simulation output loads too: a `.gro` file stands on its own, while `.xtc`, `.trr` and `.dcd` frames attach to a model that is already open and have to match its atom count. [More on supported formats](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n\n## Where the two tools differ\n\n\u003Cdiv class=\"logo-row\" style=\"display:flex;flex-wrap:wrap;align-items:center;justify-content:center;gap:1.75rem;background:#fff;border-radius:12px;padding:1.25rem 1.5rem;margin:2rem 0\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_nanome_is_different_from_vmd_image_4_ebf8fda2f1.png\" alt='VMD' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_nanome_is_different_from_vmd_image_6_ccf727e454.png\" alt='OpenEye' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_nanome_is_different_from_vmd_image_7_4a087f1a58.png\" alt='Cresset Flare' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_nanome_is_different_from_vmd_image_8_1d923cedd6.png\" alt='CDD Vault' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_nanome_is_different_from_vmd_image_9_7bde76e5db.png\" alt='KNIME' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003C\u002Fdiv>\n\n\n\n![Two colleagues wearing ultra-thin VR headsets examine the same ribbon-cartoon protein structure floating between them](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_nanome_is_different_from_vmd_image_1_v4s_d5722649b9.png)\n\n\n\nVMD is built for one person at a workstation, with a console for whatever trajectory analysis that person wants to write. Nanome wraps a few layers around the viewer: other people standing in the same 3D scene, headset rendering, and an AI that takes requests in words.\n\n\u003Ctable class=\"table\">\n  \u003Cthead>\n    \u003Ctr>\u003Cth>Tool\u003C\u002Fth>\u003Cth>Strengths\u003C\u002Fth>\u003Cth>What Nanome adds\u003C\u002Fth>\u003C\u002Ftr>\n  \u003C\u002Fthead>\n  \u003Ctbody>\n    \u003Ctr>\u003Ctd>VMD\u003C\u002Ftd>\u003Ctd>Free trajectory visualization and measurement, scripted analysis, comfortable with very large simulation datasets\u003C\u002Ftd>\u003Ctd>Shared sessions, headset rendering, and MARA running tools on the same structures from plain-English requests\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Nanome\u003C\u002Ftd>\u003Ctd>Multi-user review in XR and the browser, the MARA copilot, \u003Ca href=\"https:\u002F\u002Fnanome.ai\u002Fintegrations\">300+ integrated tools\u003C\u002Fa>\u003C\u002Ftd>\u003Ctd>The platform under discussion here\u003C\u002Ftd>\u003C\u002Ftr>\n  \u003C\u002Ftbody>\n\u003C\u002Ftable>\n\nFour things separate them in practice.\n\n**Several people in one scene.** A Nanome session holds more than one scientist at a time, each with their own vantage point, pointing at the same residue. A peer-reviewed evaluation of the platform puts the ceiling at 20 participants, with roughly 6 as the comfortable number once everyone is actively building.\n\n**Headsets, and a browser for everyone else.** Nanome [runs on Apple Vision Pro, Meta Quest, Pico Neo and HTC Vive Focus 3, with a Windows desktop build and a browser app that wants no headset at all](https:\u002F\u002Fnanome.ai\u002Fsetup). Inside a headset you stand beside a binding pocket at arm's length rather than orbiting it with a mouse.\n\n**MARA, the copilot.** Ask for an analysis in ordinary words and MARA runs it. The built-in library spans 26 categories: docking with Smina and DiffDock-L, co-folding and structure prediction across several engines including [AlphaFold 3](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41586-024-07487-w) and [Boltz-2](https:\u002F\u002Fgithub.com\u002Fjwohlwend\u002Fboltz), electrostatics through APBS, ADMET and toxicity prediction, ProteinMPNN for sequence design and ANARCI for antibody numbering and CDR loops, [RFdiffusion3 (beta)](https:\u002F\u002Fwww.ipd.uw.edu\u002F2025\u002F12\u002Frfdiffusion3-now-available\u002F) for de novo binders, and molecular dynamics trajectory analysis on whatever is loaded. Each run lists its inputs and its outputs, so a reviewer can retrace it.\n\n**Playback with company.** Nanome plays back the frame trajectories a simulation produces, with a 2000-frame ceiling on each one. Nanome is the only one of the two where the trajectory plays in a live multi-user session that a colleague joins from a standalone headset or a browser tab, with nothing to install.\n\nThat last point decided a two-year standoff at Resonac. Its computational group had GROMACS trajectories showing which fatty-alcohol additives wrapped a vitamin C derivative in a tight micelle and which stacked into flat lamellar layers, and its experimental group, reasonably, wanted more than plots and cross-sections before changing an established protocol. Both groups opened the same trajectories in Nanome, walked around the aggregates together, and agreed inside an afternoon. A six-month experimental iteration cycle came down to two or three days. The peer-reviewed [Kingsley et al. 2019 study](https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.jmgm.2019.03.010), written with a Novartis team, described the same communication gap years earlier, and it sits alongside Nanome's other [published work](https:\u002F\u002Fnanome.ai\u002Fpublications).\n\nNanome slots into a stack rather than replacing one. There's a REST API, MCP servers and a Nanome Claude Code Skill for scripted access, plus connections to [Schrödinger LiveDesign](https:\u002F\u002Fnanome.ai\u002Fintegrations), [KNIME](https:\u002F\u002Fwww.knime.com), Jupyter, [CDD Vault](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fcollaborative-drug-discovery-and-nanome-partnership-announcement), Cresset Flare, [OpenEye (Cadence)](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fview-your-openeyecadence-data-with-fresh-eyes-using-nanome-xr) and the OpenFold Consortium.\n\n## When VMD is the right pick\n\n![A researcher studies overlaid molecular dynamics trajectory frames rendered in space-filling style on a large curved display in a quiet research office.](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_nanome_is_different_from_vmd_image_2_5cf05a907b.png)\n\n\n\nA group whose week is mostly simulation, running long jobs and writing code to slice the output a particular way, is well served by VMD. It costs nothing, it stays quick on enormous datasets, and its scripting is the reason it has held its place in molecular dynamics for decades. Plenty of scientists run both, with VMD doing the heavy trajectory scripting and Nanome carrying the group review and the analysis requests that are quicker to say than to script.\n\nThat pairing shows up across the rest of this comparison set. The [PyMOL write-up](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fhow-nanome-is-different-from-pymol) is the hub for all of them, and there are separate pieces on [BIOVIA Discovery Studio](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fhow-nanome-is-different-from-biovia-discovery-studio) and on [MOE](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fhow-nanome-is-different-from-moe), which happens to be a listed Nanome integration as well as a comparison.\n\n## Deployment and access\n\n![A flat vector diagram showing a building outline containing a molecule, an AI chip, and a document, all enclosed within a bold firewall boundary with a lock icon.](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_nanome_is_different_from_vmd_image_3_bcc1b9b044.png)\n\n\n\nEnterprise deployments, MARA included, [stay inside your own network, as single-tenant cloud or fully on-prem](https:\u002F\u002Fnanome.ai\u002Fmara-enterprise). Structures, sequences and tool outputs never cross that boundary. A collaborator with no headset and no admin rights on their laptop still joins the same session from a browser tab.\n\nThe [case studies](https:\u002F\u002Fnanome.ai\u002Fcase-studies) cover what those sessions looked like on live programs, Resonac's among them.\n\n## FAQ\n\n**What are good alternatives to VMD?**\nNanome is one, for groups that want several people in the structure at once and immersive 3D on a headset. It opens the same trajectory files, plays them back, and adds the MARA copilot for analysis requested in words. Where the work is trajectory analysis written and re-run as scripts, VMD itself is still a sound answer.\n\n**Can Nanome handle molecular dynamics trajectories like VMD?**\nNanome plays back frame trajectories: `.gro` on its own, and `.xtc`, `.trr` or `.dcd` attached to a structure already loaded, with atom counts matching. MARA can then run trajectory analysis on top. VMD was built around molecular dynamics and its scripting goes deeper, so heavy custom simulation analysis often stays there.\n\n**Is Nanome free like VMD?**\nVMD is free for non-commercial and internal use under its University of Illinois license. Nanome is a commercial platform, and its enterprise deployments run inside a customer's own network.\n\n**Can I use Nanome without a headset?**\nYes. The browser web app needs no headset and no install, and someone using it shares a session with colleagues who are wearing one. There's also a Windows desktop build.\n\n**What file formats does Nanome read?**\nStructures come in as PDB, mmCIF, SDF, MOL and MOL2, XYZ, PQR, SMILES, and `.pdbqt` from AutoDock (converted to PDB, charges dropped). Session files import for viewing: Maestro `.mae` and `.maegz`, `.moe`, and PyMOL `.pse`. Trajectories use the frame-trajectory path: `.gro` standalone, with `.xtc`, `.trr` and `.dcd` attaching to a model already loaded. A `.dx` electrostatic map overlays a loaded model. Export runs to PDB, SDF or SMILES, a single frame at a time, and mmCIF, MAE, MOE and PSE are read-only. Electron density maps (CCP4, MRC, DSN6) and native LAMMPS trajectories aren't supported. [Full list](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n","2026-07-15T01:23:51.621Z","2026-09-10T16:00:12.145Z","2026-09-10T16:00:12.055Z","A fair VMD alternative guide: where VMD leads on MD trajectories and scripting, and where Nanome adds multiplayer, XR, and its MARA AI copilot.","alternatives to VMD, VMD alternative, VMD trajectory visualization, molecular dynamics visualization, Nanome, MARA, XR molecular visualization","how-nanome-is-different-from-vmd",{"id":199,"attributes":200},145,{"title":201,"content":202,"createdAt":203,"updatedAt":204,"publishedAt":205,"date":172,"description":206,"keywords":207,"slug":208,"category":17},"How to run an SAR meeting on a 3D structure","An SAR meeting is the recurring project review where chemists, biologists and modelers read the structure-activity data off the latest round of compounds and settle what to make next. It runs well when the structure under discussion is in front of everyone at once, when the questions raised in the room get answered inside the hour, and when the reasoning is still legible to whoever wasn't there. Nanome, a collaborative molecular visualization and drug discovery platform, holds that meeting as a shared 3D session: the compounds are laid out beforehand as [scenes](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_v2\u002Fscenespanel), saved views of the real structure that step through like an agenda, and colleagues join from a browser tab or a headset. MARA, the AI copilot inside Nanome, docks an analog or returns ADMET numbers while the discussion is still live.\n\n## What an SAR meeting has to produce\n\nTwo things leave the room. One is a decision: a short list of compounds to make, in some order, with a name against each. The other is the reasoning behind that list, in a form that still holds up 6 weeks later when the assay data lands and the group has to work out whether the hypothesis was wrong or the chemistry was.\n\nMost of the argument in between is spatial. A substituent gains potency because it reaches into a subpocket, or loses it because it bumps a backbone carbonyl at an angle a flat depiction flattens away. Getting the contacts themselves right is a job of its own, and [tools for visualizing protein-ligand interactions](https:\u002F\u002Fnanome.ai\u002Fblog\u002Ftools-for-visualizing-protein-ligand-interactions) works through the options.\n\n## The shape of the meeting\n\n![Two colleagues wearing ultra-thin VR headsets examine the same solid protein surface model with a small ligand visible in its binding pocket floating between them](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_to_run_an_sar_meeting_image_1_v4_a2494db83a.png)\n\n\n\n1. **Scenes built beforehand.** Someone sets up one scene per compound, or per hypothesis if 2 analogs are making the same argument. Scenes shipped with Nanome 2.0, and each carries its own point of view from v2.1.1, so the camera is already where the discussion needs it when that scene comes up. Dragging reorders them, which is usually how the agenda gets its final shape 10 minutes before the call.\n\n2. **The room.** The session opens in the web app or in a headset, and both kinds of participant are in the same workspace. Colored dots show which scene each person is looking at. Spotlight Mode puts the presenter's view on the screens of everyone following, so a group that has drifted apart can be pulled back to one perspective without anyone reading out camera coordinates.\n\n3. **Walking the scenes.** The object on screen is the live structure rather than a picture of one, so a question about the far face of the pocket is answered by turning it. Someone who wants to lean in and check a distance can do that while the conversation carries on.\n\n4. **The question nobody prepared for.** MARA takes it in plain English inside the session: dock the new analog with Smina or DiffDock-L, run ADMET across the short series, compute the interactions around a pose. Every run leaves its own trail: the tool that fired, the structure it was pointed at, and the numbers it sent back. A claim in the meeting notes can be walked back to the job that produced it.\n\n5. **What leaves with everyone.** A shareable workspace link, whose scenes hold the views the group settled on, and a PowerPoint slide generated from a scene when a slide has to go into a deck anyway.\n\n## How teams run it today\n\n\u003Cdiv class=\"logo-row\" style=\"display:flex;flex-wrap:wrap;align-items:center;justify-content:center;gap:1.75rem;background:#fff;border-radius:12px;padding:1.25rem 1.5rem;margin:2rem 0\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_to_run_an_sar_meeting_image_4_0566b90fcd.png\" alt='PyMOL' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_to_run_an_sar_meeting_image_5_162d8b37da.png\" alt='Schrödinger Maestro' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_to_run_an_sar_meeting_image_6_4aea2efd48.png\" alt='MOE' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_to_run_an_sar_meeting_image_7_57442cc47d.png\" alt='Mol*' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_to_run_an_sar_meeting_image_8_7e752edfc4.png\" alt='UCSF ChimeraX' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003C\u002Fdiv>\n\n\n\n\u003Ctable class=\"table\">\n  \u003Cthead>\n    \u003Ctr>\u003Cth>How the meeting runs\u003C\u002Fth>\u003Cth>What it's good at\u003C\u002Fth>\u003Cth>When someone asks for a different angle\u003C\u002Fth>\u003C\u002Ftr>\n  \u003C\u002Fthead>\n  \u003Ctbody>\n    \u003Ctr>\u003Ctd>A deck of rendered figures, often \u003Ca href=\"https:\u002F\u002Fpymol.org\">PyMOL\u003C\u002Fa> into PowerPoint\u003C\u002Ftd>\u003Ctd>Portable and archivable, opens on any device, easy to circulate before and after\u003C\u002Ftd>\u003Ctd>The camera was fixed at render time, so the question waits for the next round of figures\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>A screen-share out of \u003Ca href=\"https:\u002F\u002Fwww.schrodinger.com\u002Fplatform\u002Fproducts\u002Fmaestro\u002F\">Maestro\u003C\u002Fa> or \u003Ca href=\"https:\u002F\u002Fwww.chemcomp.com\">MOE\u003C\u002Fa>\u003C\u002Ftd>\u003Ctd>Full modeling power live in the meeting, with real measurements on demand\u003C\u002Ftd>\u003Ctd>One person drives, and everyone else describes in words what they want moved\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>A \u003Ca href=\"https:\u002F\u002Fmolstar.org\">Mol*\u003C\u002Fa> viewer embedded in a page or sent as a link\u003C\u002Ftd>\u003Ctd>Opens in any browser with nothing installed and no account, ideal for a public PDB entry\u003C\u002Ftd>\u003Ctd>Each viewer turns it privately, so the shared reference frame quietly disappears\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>A rendered animation, for example a \u003Ca href=\"https:\u002F\u002Fwww.rbvi.ucsf.edu\u002Fchimerax\u002F\">ChimeraX\u003C\u002Fa> \u003Ccode>movie\u003C\u002Fcode> script\u003C\u002Ftd>\u003Ctd>Plays anywhere, and the camera path can be composed with real care beforehand\u003C\u002Ftd>\u003Ctd>It plays the same way every time, so a new angle means a new render\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Scenes in a shared Nanome session\u003C\u002Ftd>\u003Ctd>Everyone holds one live structure between them, from a browser tab or a headset\u003C\u002Ftd>\u003Ctd>Anyone turns it for themselves, and the scene's saved point of view brings the group back\u003C\u002Ftd>\u003C\u002Ftr>\n  \u003C\u002Ftbody>\n\u003C\u002Ftable>\n\n## Where Nanome fits\n\n![Two colleagues discuss a protein surface structure on a wall-mounted display during a post-meeting review.](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_to_run_an_sar_meeting_image_2_630fc89cc1.png)\n\n\n\nNanome supplies the room and the thing that leaves it. Shareable workspaces arrived with the web app in v2.5, along with projects and permissions, so the meeting has a link and an access list. A biologist who missed the call opens the same scenes in a browser afterwards and sees the pocket at the orientation the argument settled on. A file plus a paragraph of setup notes rarely carries that much.\n\nThere's evidence this shape of meeting produces chemistry. A 2019 paper written jointly by a Novartis GNF team and Nanome, published in the Journal of Molecular Graphics and Modelling, documents 4 chemists exploring macrocyclization options together on an active compound inside the software. The series they went on to synthesize held its activity and came back with improved PK properties ([doi.org\u002F10.1016\u002Fj.jmgm.2019.03.010](https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.jmgm.2019.03.010)).\n\nGetting the compounds in takes no conversion step. PDB, mmCIF, SDF, MOL and MOL2, XYZ, PQR, SMILES and PDBQT all load, as do Maestro `.mae` and `.maegz` files, MOE `.moe` files and PyMOL `.pse` sessions, though those last 4 travel in one direction only. What Nanome writes back out is PDB, SDF or SMILES, one frame at a time. [More on supported formats](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n\nThe wider case for running med chem discussions this way sits in [VR for medicinal chemists](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fvr-for-medicinal-chemists), and the mechanics of building the scenes are in [what a modern molecular presentation looks like](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fwhat-a-modern-molecular-presentation-looks-like).\n\n## When a deck and a call is the right pick\n\nA 20-person portfolio review is a deck. Coverage is the job there, one slide per program, and nobody is turning a pocket while 19 people wait for the next agenda item.\n\nAn external partner with no account is a link. For a public PDB entry, a Mol* viewer embedded in a page wins outright on the thing that matters most in that moment: the recipient clicks, the structure is there, nothing installed and nobody provisioned. ChimeraX has a genuine answer here too. Its `movie` command composes a camera path and writes a file that plays on any laptop in any conference room, with no session to join.\n\nAnd some meetings turn on a number rather than a shape. If the decision hangs on a DMPK column or a potency ranking, a table settles it faster than any 3D view will. Write-ups of projects where the 3D did carry the decision sit at [nanome.ai\u002Fcase-studies](https:\u002F\u002Fnanome.ai\u002Fcase-studies).\n\n## FAQ\n\n**What is an SAR meeting?**\nA recurring project review where the team reads structure-activity data from the latest compounds and decides what to make next. Chemists, biologists and modelers are usually all in it. Two things come out: a synthesis list, and the reasoning that justifies it well enough to revisit when the next assay data arrives.\n\n**How do you run an SAR meeting with a remote team?**\nEveryone joins one shared session rather than watching a screen-share. In Nanome the compounds are already laid out as scenes with their own saved camera angles, colored dots show which scene each person is on, and Spotlight Mode lets the presenter put their view on everyone else's screen. The workspace link then carries the whole meeting to anyone in a time zone that couldn't make it.\n\n**Can biologists take part without a headset?**\nYes. The Nanome web app runs in a browser tab with nothing to install, and a colleague in a headset and a colleague on a laptop share the same session and the same molecule. Mixed rooms like that are the normal case for an SAR review: the modeler in the headset and the biologist on a laptop are in one session, looking at one pocket.\n\n**What comes out of the meeting?**\nA decision and a record of it. In practice that means a shareable workspace holding the scenes and the views the group agreed on, a PowerPoint slide generated from a scene when the deck still needs one, and a log of the tools MARA ran with their inputs and results, so a number quoted in the follow-up email can be traced to the run that produced it.\n","2026-08-27T18:19:50.815Z","2026-09-10T16:00:16.227Z","2026-09-10T16:00:16.106Z","An SAR meeting decides what to make next. How to run one on a live 3D structure: scenes as the agenda, a shared session, and a record that outlives the room.","SAR meeting, structure activity relationship meeting, medicinal chemistry design review, run an SAR meeting, design review best practices","how-to-run-an-sar-meeting",{"id":210,"attributes":211},95,{"title":212,"content":213,"createdAt":214,"updatedAt":215,"publishedAt":216,"date":217,"description":218,"keywords":219,"slug":220,"category":17},"How Nanome is different from Schrödinger Maestro","Nanome is a collaborative molecular visualization and drug discovery platform that runs on the web and in XR headsets, and it ships with an [AI copilot called MARA](https:\u002F\u002Fnanome.ai\u002Fmara) that you drive in plain English. [Schrödinger](https:\u002F\u002Fwww.schrodinger.com) [Maestro](https:\u002F\u002Fwww.schrodinger.com\u002Fplatform\u002Fproducts\u002Fmaestro\u002F) is a mature desktop computational chemistry suite that packs docking, free-energy calculations, and modeling into one deep environment. Both open the same PDB and SDF files, so the difference shows up in how you work: Maestro gives one scientist a powerful desktop cockpit, and Nanome adds real-time multiplayer sessions, native XR, and a copilot that runs the tools on request. Nanome also integrates with [Schrödinger LiveDesign](https:\u002F\u002Fnanome.ai\u002Fintegrations), so it fits next to that stack rather than trying to swap it out.\n\nThe phrase \"Schrödinger Maestro alternative\" covers two fairly different needs: a deeper computational engine, or a way to get more people around the same structure at once. Maestro and Nanome answer different halves of that.\n\n## Where Maestro is strong\n\n\u003Cdiv class=\"logo-row\" style=\"display:flex;flex-wrap:wrap;align-items:center;justify-content:center;gap:1.75rem;background:#fff;border-radius:12px;padding:1.25rem 1.5rem;margin:2rem 0\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_nanome_is_different_from_schrodinger_maestro_image_5_262e193c2a.png\" alt='MOE (Molecular Operating Environment)' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_nanome_is_different_from_schrodinger_maestro_image_7_4f867293da.png\" alt='OpenEye' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_nanome_is_different_from_schrodinger_maestro_image_9_8644c91e16.png\" alt='CDD Vault' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003C\u002Fdiv>\n\n\n\n![A computational chemist reviews a molecular surface structure on a large display in a modern research office.](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_nanome_is_different_from_schrodinger_maestro_image_1_368b7e7d7b.png)\n\n\n\nMaestro has grown over decades into a broad, tightly integrated suite. It's the front end to a large family of Schrödinger tools: precise docking, physics-based free-energy perturbation, homology modeling, quantum mechanics, and more, all wired together with validated workflows. For a comp-chem specialist who lives in that ecosystem day to day, it's a solid pick and often the right one.\n\n[MOE (Molecular Operating Environment)](https:\u002F\u002Fwww.chemcomp.com) and [BIOVIA Discovery Studio](https:\u002F\u002Fwww.3ds.com\u002Fproducts\u002Fbiovia\u002Fdiscovery-studio) occupy nearby ground: full desktop suites with deep modeling toolkits, driven through a rich GUI or a scripting layer, and built around one scientist at a keyboard. Those two have their own write-ups, [Nanome next to MOE](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fhow-nanome-is-different-from-moe) and [Nanome next to Discovery Studio](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fhow-nanome-is-different-from-biovia-discovery-studio). PyMOL belongs to Schrödinger as well, and [that comparison](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fhow-nanome-is-different-from-pymol) runs along different lines again.\n\nWhen files already live in one of these tools, Nanome meets them where they are. It opens Maestro's own `.mae` and `.maegz`, which is what the LiveDesign gadget hands over, plus MOE's `.moe`. The everyday structure formats load too: PDB, mmCIF, SDF, MOL and MOL2, SMILES, XYZ, PQR, and `.pdbqt` from the AutoDock side. [More on supported formats](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n\n## Where Nanome is different\n\n![Two colleagues wearing ultra-thin VR headsets examine the same ribbon-cartoon protein structure floating between them](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_nanome_is_different_from_schrodinger_maestro_image_2_v4s_0602f95dda.png)\n\n\n\nNanome starts from those same structures and pushes in a different direction: shared 3D space, immersive hardware, and a copilot with the scientific tools already wired in.\n\n\u003Ctable class=\"table\">\n  \u003Cthead>\n    \u003Ctr>\u003Cth>Maestro is good at\u003C\u002Fth>\u003Cth>Where Nanome is different\u003C\u002Fth>\u003C\u002Ftr>\n  \u003C\u002Fthead>\n  \u003Ctbody>\n    \u003Ctr>\u003Ctd>A deep, validated desktop comp-chem suite\u003C\u002Ftd>\u003Ctd>Real-time multiplayer sessions where several people edit one structure together\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Precise physics-based docking and free-energy work\u003C\u002Ftd>\u003Ctd>\u003Ca href=\"https:\u002F\u002Fnanome.ai\u002Fsetup\">Native XR on Apple Vision Pro, Meta Quest, Pico Neo, and HTC Vive Focus 3\u003C\u002Fa>\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>A powerful single-user desktop environment\u003C\u002Ftd>\u003Ctd>A browser web app that needs no headset at all\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Menu and script-driven workflows\u003C\u002Ftd>\u003Ctd>Plain-English requests that MARA turns into runs across \u003Ca href=\"https:\u002F\u002Fnanome.ai\u002Fintegrations\">300+ scientific tools\u003C\u002Fa>\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>A self-contained Schrödinger ecosystem\u003C\u002Ftd>\u003Ctd>Integrates with Schrödinger LiveDesign, plus \u003Ca href=\"https:\u002F\u002Fnanome.ai\u002Fblog\u002Fcollaborative-drug-discovery-and-nanome-partnership-announcement\">CDD Vault\u003C\u002Fa>, \u003Ca href=\"https:\u002F\u002Fwww.knime.com\">KNIME\u003C\u002Fa>, Jupyter, and \u003Ca href=\"https:\u002F\u002Fwww.eyesopen.com\">OpenEye\u003C\u002Fa>\u003C\u002Ftd>\u003C\u002Ftr>\n  \u003C\u002Ftbody>\n\u003C\u002Ftable>\n\nThe multiplayer piece matters more than it sounds on paper. Two chemists in different cities can stand around the same protein, point into the same pocket, and rotate a ligand together in real time, each with their own viewpoint and their own hands on the structure.\n\nThen there's the copilot. A plain-English request is enough, and MARA runs the job: docking through Smina and DiffDock-L, co-folding and structure prediction across [AlphaFold 3](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41586-024-07487-w), [Boltz-2](https:\u002F\u002Fgithub.com\u002Fjwohlwend\u002Fboltz), and OpenFold3, electrostatics with APBS, ADMET and toxicity models, sequence design with ProteinMPNN, de novo binder design with RFdiffusion3 (beta), and ANARCI for antibody numbering and CDR loops. Each run reports the tool it called, the inputs it got, and what came back, so a colleague can retrace the reasoning later.\n\nMotion comes across as well. Nanome loads and plays back MD trajectories. A `.gro` file loads on its own; `.xtc`, `.trr`, and `.dcd` attach to a model already in the workspace and have to match its atom count. Frame-trajectory playback runs to a 2000-frame cap, and surfaces switch off while a trajectory plays. [More on supported formats](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n\nThe LiveDesign connection is the practical part for a Schrödinger shop. The gadget hands a LiveReport's structures straight into Nanome, so a team can bring today's series into a shared immersive session without stepping outside the pipeline they already run. Nanome also connects to Cresset Flare, CDD Vault, KNIME, Jupyter, [the OpenFold Consortium](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fnanome-joins-the-openfold-consortium), and OpenEye (Cadence), with an open REST API and MCP servers underneath.\n\nA Novartis GNF project shows what that looks like on a live series. Four chemists took an active compound into Nanome to weigh macrocyclization options, and people new to the software were picking out candidate sites within a few minutes. The macrocycle series they went on to synthesize held its activity and came back with improved PK properties. The write-up is [Kingsley et al. (2019)](https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.jmgm.2019.03.010) in the Journal of Molecular Graphics and Modelling, co-authored by Novartis GNF and Nanome scientists. More of that work sits at [nanome.ai\u002Fcase-studies](https:\u002F\u002Fnanome.ai\u002Fcase-studies).\n\n## When Maestro is the right pick\n\n![A researcher wearing an ultra-thin VR headset studies a space-filling protein model held close to the chest](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_nanome_is_different_from_schrodinger_maestro_image_3_v4s_28939a6843.png)\n\n\n\nWork that centers on physics-based free-energy calculations, high-precision docking campaigns, or the deep modeling Schrödinger has spent years validating is Maestro's home ground, and MOE and Discovery Studio play in the same range. A specialist running rigorous quantitative predictions can do that job without multiplayer or a headset anywhere in the picture.\n\nNanome fits when several people have to decide together, when seeing a structure at true 3D scale changes what gets noticed, or when a plain-English request should kick off the docking run. Plenty of groups run both: Maestro and LiveDesign for the rigorous computation, Nanome for the shared exploration and the group review around it.\n\n## FAQ\n\n**Is Nanome an alternative to Schrödinger Maestro?**\nIt overlaps on visualization and diverges from there. Nanome opens the same PDB and SDF structures, then adds real-time multiplayer collaboration, native XR, and MARA. Because it connects to Schrödinger LiveDesign, many teams run it alongside Maestro.\n\n**Does Nanome integrate with Schrödinger?**\nYes, through Schrödinger LiveDesign. The gadget carries a LiveReport's structures into a shared immersive session while the existing pipeline keeps running. Nanome also connects to CDD Vault, KNIME, Jupyter, Cresset Flare, and OpenEye.\n\n**Is a VR headset required?**\nNo. Nanome runs as a browser web app and on Windows desktop with nothing on your head. XR on Apple Vision Pro, Meta Quest, Pico Neo, and HTC Vive Focus 3 is there for the moments immersive 3D helps.\n\n**Which file formats does Nanome open?**\nStructures import as PDB and `.ent`, mmCIF and its variants (`.cif`, `.mmcif`, `.mcif`, `.bcif`), SDF, MOL and MOL2, SMILES, XYZ, PQR, and `.pdbqt`. Vendor and session files import too: Maestro `.mae` and `.maegz`, MOE `.moe`, and PyMOL `.pse`. Trajectories cover `.gro`, `.xtc`, `.trr`, and `.dcd`, while a `.dx` electrostatic map overlays a model that is already loaded. Export is PDB, SDF, or SMILES, single frame, which makes mmCIF, MAE, MOE, and PSE read-only. Full detail lives at [docs.nanome.ai](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n\n**What can MARA run?**\nMARA can call 300+ built-in tools across 26 categories: docking, co-folding and structure prediction, electrostatics, ADMET and toxicity, cheminformatics, antibody numbering and CDR definition, de novo binder design, and MD trajectory analysis. Each answer names the tools it called and what they returned.\n","2026-07-15T01:23:51.485Z","2026-09-08T16:00:11.340Z","2026-09-08T16:00:11.286Z","2026-09-08","A Schrödinger Maestro alternative? Nanome is a collaborative, XR-native platform with an AI copilot, and it integrates with Schrödinger LiveDesign.","alternatives to Schrödinger Maestro, Schrödinger Maestro alternative, molecular modeling software, computational chemistry, Nanome, MARA, MOE, LiveDesign","how-nanome-is-different-from-schrodinger-maestro",{"id":222,"attributes":223},94,{"title":224,"content":225,"createdAt":226,"updatedAt":227,"publishedAt":228,"date":217,"description":229,"keywords":230,"slug":231,"category":17},"How Nanome is different from PyMOL","Nanome is a collaborative molecular visualization and drug discovery platform that runs on the web and in XR headsets. Inside it, an [AI copilot called MARA](https:\u002F\u002Fnanome.ai\u002Fmara) runs scientific tools from plain-English requests. [PyMOL](https:\u002F\u002Fpymol.org), now maintained by [Schrödinger](https:\u002F\u002Fwww.schrodinger.com), is a script-driven desktop viewer that's widely used for publication-quality images and precise rendering. Both open the same PDB and SDF files, so the real difference is how you work with a structure once it's loaded: PyMOL gives you one person at a keyboard, and Nanome adds real-time multiplayer sessions and native VR around the same coordinates.\n\nThey overlap on the file and separate on what happens after it opens.\n\n## Where PyMOL is strong\n\n\u003Cdiv class=\"logo-row\" style=\"display:flex;flex-wrap:wrap;align-items:center;justify-content:center;gap:1.75rem;background:#fff;border-radius:12px;padding:1.25rem 1.5rem;margin:2rem 0\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_nanome_is_different_from_pymol_image_4_1ee18a29eb.png\" alt='PyMOL' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_nanome_is_different_from_pymol_image_5_af8e919597.png\" alt='Schrödinger' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_nanome_is_different_from_pymol_image_6_c46d666ebf.png\" alt='VMD' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_nanome_is_different_from_pymol_image_8_1e81cbfd3a.png\" alt='GROMACS' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_nanome_is_different_from_pymol_image_9_0b4beab5df.png\" alt='OpenEye' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003C\u002Fdiv>\n\n\n\n![A researcher in a knit sweater reviews a ribbon-rendered protein structure on a flat monitor in a naturally lit office.](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_nanome_is_different_from_pymol_image_1_756c3796ff.png)\n\n\n\nPyMOL has been a lab staple for over 20 years. It renders beautiful ray-traced images, its scripting language gives you exact control over every atom and camera angle, and there's a deep well of community scripts and tutorials. For a figure in a paper, a reproducible rendering pipeline, or quick single-user structure inspection, it's a solid pick and often the right one.\n\nNanome imports PyMOL `.pse` sessions, so a scene built there can open in a shared workspace. Which PyMOL session versions are validated isn't documented, and QM\u002FMM link atoms can break the parse, so a test load is worth doing before a meeting depends on the file. Traffic runs one way. Nanome writes PDB, SDF or SMILES, single frame, and there's no `.pse` writer. [What imports and what exports](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n\n[VMD](https:\u002F\u002Fwww.ks.uiuc.edu\u002FResearch\u002Fvmd\u002F) and [UCSF ChimeraX](https:\u002F\u002Fwww.rbvi.ucsf.edu\u002Fchimerax\u002F) sit in the same neighborhood. VMD is strong for molecular dynamics trajectory work, and ChimeraX is a modern desktop viewer with excellent analysis and rendering. All 3 are desktop, mostly single-user, and driven by a GUI or a command line.\n\nNanome loads and plays back MD trajectories. A `.gro` file loads on its own; `.xtc`, `.trr`, and `.dcd` attach to a model already in the workspace and have to match its atom count. Frame-trajectory playback carries a 2000-frame cap, and surfaces switch off while a trajectory plays. [More on supported formats](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n\n## Where Nanome is different\n\n![A researcher wearing a slim VR headset examines a solid protein surface model floating at chest height in an open studio.](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_nanome_is_different_from_pymol_image_2_r3_05d587c82e.png)\n\n\n\nNanome starts from the same files and goes in a different direction: shared 3D space, immersive hardware, and an AI copilot.\n\n\u003Ctable class=\"table\">\n  \u003Cthead>\n    \u003Ctr>\u003Cth>PyMOL is good at\u003C\u002Fth>\u003Cth>Where Nanome is different\u003C\u002Fth>\u003C\u002Ftr>\n  \u003C\u002Fthead>\n  \u003Ctbody>\n    \u003Ctr>\u003Ctd>Publication-quality ray-traced images\u003C\u002Ftd>\u003Ctd>Real-time multiplayer sessions where several people edit one structure together\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Precise scripting control over rendering\u003C\u002Ftd>\u003Ctd>\u003Ca href=\"https:\u002F\u002Fnanome.ai\u002Fsetup\">Native XR and VR on Meta Quest and Apple Vision Pro, plus HTC Vive Focus 3 and Pico Neo\u003C\u002Fa>\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Single-user desktop inspection\u003C\u002Ftd>\u003Ctd>A browser app that needs no headset, and a Windows desktop build\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Community scripts for repeated tasks\u003C\u002Ftd>\u003Ctd>MARA, an AI copilot that runs \u003Ca href=\"https:\u002F\u002Fnanome.ai\u002Fintegrations\">300+ scientific tools\u003C\u002Fa> from plain-English requests\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Local files on your machine\u003C\u002Ftd>\u003Ctd>Structures fetched straight from RCSB PDB, PubChem, and DrugBank into the session\u003C\u002Ftd>\u003C\u002Ftr>\n  \u003C\u002Ftbody>\n\u003C\u002Ftable>\n\nThe multiplayer part matters more than it sounds. In Nanome, 2 chemists in different cities can stand around the same protein in VR, point at a pocket, and rotate the ligand together in real time. That's a different kind of review than screen-sharing a static render.\n\nPyMOL renders the figure that goes in the paper, and Nanome holds the object the group turns during the talk. Most teams do both, which is the argument in [what a modern molecular presentation looks like](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fwhat-a-modern-molecular-presentation-looks-like).\n\nThen there's MARA. You describe the job in plain English and it runs, so the compute lands next to the structure instead of in another window. It covers docking (Smina, DiffDock-L), electrostatics with APBS, ADMET prediction, structure prediction and co-folding with [AlphaFold 3](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41586-024-07487-w) and [Boltz-2](https:\u002F\u002Fgithub.com\u002Fjwohlwend\u002Fboltz), antibody numbering and CDR definition with ANARCI, sequence design with ProteinMPNN, and de novo binder design with RFdiffusion3 (beta). Each run reports the tool it called, the inputs it used, and what came back, so the work is checkable.\n\nNanome fits into the stack around it. Since Schrödinger maintains PyMOL, the [LiveDesign](https:\u002F\u002Fnanome.ai\u002Fintegrations) connection is the relevant one: Nanome imports Maestro `.mae` and `.maegz`, the format LiveDesign uses to hand structures off. It also connects to [OpenEye (Cadence)](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fview-your-openeyecadence-data-with-fresh-eyes-using-nanome-xr), Cresset Flare, [CDD Vault](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fcollaborative-drug-discovery-and-nanome-partnership-announcement), [KNIME](https:\u002F\u002Fwww.knime.com), and Jupyter. The REST API and MCP servers are open, for whatever isn't on that list.\n\nThe comparison has a published number behind it. The 2019 platform paper ([Kingsley et al., J. Mol. Graph. Model. 89, 234-241](https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.jmgm.2019.03.010)), written with a Novartis GNF team, timed Nanome's surface generation against PyMOL, Discovery Studio, Chimera and ChimeraX and found it several times faster, while holding 90+ frames per second across two eye views. That was 2019 hardware and 2019 code on every side, so treat the number as a floor. The [publications](https:\u002F\u002Fnanome.ai\u002Fpublications) page has the rest.\n\n## When PyMOL is the right pick\n\n![A researcher at a bare desk studies a ribbon-cartoon protein structure displayed on a flat monitor in a quietly lit office.](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_nanome_is_different_from_pymol_image_3_r3_176d352f0f.png)\n\n\n\nIf the deliverable is a polished static figure, a scripted rendering pipeline under version control, or fast single-user inspection on a laptop, PyMOL (or ChimeraX, or VMD) does that well. A clean image of a binding site needs no multiplayer and no headset.\n\nNanome covers the other half of the week: collaborative review, structures where true 3D scale changes what you notice, and jobs that are quicker to describe than to script. Plenty of groups run both, PyMOL for the final figure and Nanome for the exploration and the group review behind it. The same split shows up against [BIOVIA Discovery Studio](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fhow-nanome-is-different-from-biovia-discovery-studio) and [CCG MOE](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fhow-nanome-is-different-from-moe), where MOE also happens to be a listed Nanome integration. The [case studies](https:\u002F\u002Fnanome.ai\u002Fcase-studies) show what that collaborative half looked like on live programs.\n\n## FAQ\n\n**Is Nanome an alternative to PyMOL?**\nFor part of the job, yes. It opens the same structures and covers visualization, then adds live multi-user sessions, native XR and VR, and MARA for the compute. Some groups move a slice of their PyMOL workflow across, and others keep both and use each where it's stronger.\n\n**What file formats does Nanome support?**\nStructures import as PDB (`.pdb`, `.ent`), mmCIF (`.cif`, `.mmcif`, `.bcif`), SDF (`.sdf`, `.mol`), MOL2, SMILES, XYZ, PQR, and PDBQT. Vendor and session files come in too: PyMOL `.pse`, Maestro `.mae` and `.maegz`, and MOE `.moe`. MD trajectories arrive as `.gro` standalone, or as `.xtc`, `.trr` and `.dcd` attached to an open model with a matching atom count, and `.dx` maps overlay a loaded model. Export is narrower: PDB, SDF or SMILES, single frame, which makes mmCIF, MAE, MOE and PSE import-only. [More on supported formats](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n\n**Can I use Nanome without a headset?**\nYes. It runs in a browser and on Windows desktop. Headset support covers Meta Quest and Apple Vision Pro, plus HTC Vive Focus 3 and Pico Neo, for when immersive 3D is the point.\n\n**What can MARA actually run?**\n300+ tools across 26 categories, called from plain-English requests: docking, co-folding, electrostatics, ADMET and toxicity prediction, antibody numbering and CDR definition, de novo binder design, structure prediction, cheminformatics, and MD trajectory analysis. Each answer names the tool it used and shows what came back.\n","2026-07-15T01:23:51.388Z","2026-09-08T16:00:09.619Z","2026-09-08T16:00:09.567Z","Nanome is a collaborative, XR-native alternative to PyMOL with an AI copilot that runs docking and structure tools from plain English.","alternatives to PyMOL, best alternatives to PyMOL for molecular visualization, PyMOL alternative, molecular visualization software, Nanome, MARA, ChimeraX, VMD","how-nanome-is-different-from-pymol",{"id":233,"attributes":234},92,{"title":235,"content":236,"createdAt":237,"updatedAt":238,"publishedAt":239,"date":217,"description":240,"keywords":241,"slug":242,"category":17},"How Nanome is different from BIOVIA Discovery Studio","Nanome is a collaborative molecular visualization and drug discovery platform that runs across a browser web app and XR headsets. It also carries [MARA, an AI copilot](https:\u002F\u002Fnanome.ai\u002Fmara) that runs scientific tools from plain-English requests. [BIOVIA Discovery Studio](https:\u002F\u002Fwww.3ds.com\u002Fproducts\u002Fbiovia\u002Fdiscovery-studio) is a broad commercial life-science modeling suite for simulation, protein modeling, and structure-based design on the desktop. If you're weighing a Discovery Studio alternative, the two overlap in some places and part ways in others, so the right pick depends on how your team wants to work.\n\nThe differences gather in three spots: who can stand in the structure with you, what the software runs on, and how an analysis gets requested.\n\n## Where Nanome is different\n\n![Two colleagues wearing ultra-thin VR headsets examine the same ribbon-cartoon protein structure floating between them](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_nanome_is_different_from_biovia_discovery_studio_image_1_v4s_7d7d7e782f.png)\n\n\n\nNanome puts several people inside the same 3D structure at once. You and a colleague can grab the same protein, rotate it, and point at the same binding pocket in real time, whether you share an office or sit on different continents. Discovery Studio installs on a workstation, and a session belongs to whoever is at that keyboard.\n\nIt runs natively in XR. Nanome works on Apple Vision Pro, Meta Quest, Pico Neo, and HTC Vive Focus 3, plus [Windows desktop and a browser when there's no headset around](https:\u002F\u002Fnanome.ai\u002Fsetup). Walking around a molecule at arm's length reads differently from spinning it with a mouse.\n\nThen MARA. Ask for a dock or a fold prediction in ordinary words and it runs the job, then reports the tool it called, the inputs it took, and the result it returned, so anyone can check the work later. MARA reaches [300+ integrated scientific tools across 26 categories](https:\u002F\u002Fnanome.ai\u002Fintegrations): docking through Smina and DiffDock-L, folding and co-folding through [Boltz-2](https:\u002F\u002Fgithub.com\u002Fjwohlwend\u002Fboltz), AlphaFold 3, and OpenFold3, electrostatics through APBS, ADMET and toxicity models, ANARCI for numbering antibody variable domains and marking their CDR loops, ProteinMPNN for sequence design, and RFdiffusion3 (beta) for de novo binders.\n\n## Quick comparison\n\n\u003Ctable class=\"table\">\n  \u003Cthead>\n    \u003Ctr>\u003Cth>Package\u003C\u002Fth>\u003Cth>Strengths\u003C\u002Fth>\u003Cth>How the two meet\u003C\u002Fth>\u003C\u002Ftr>\n  \u003C\u002Fthead>\n  \u003Ctbody>\n    \u003Ctr>\u003Ctd>BIOVIA Discovery Studio\u003C\u002Ftd>\u003Ctd>Broad desktop suite for simulation, protein modeling, and structure-based design\u003C\u002Ftd>\u003Ctd>Nanome opens the structures it writes and puts them in a session several people can join\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Nanome\u003C\u002Ftd>\u003Ctd>Shared visualization in the browser and in XR, with MARA driving 300+ tools\u003C\u002Ftd>\u003Ctd>Plenty of groups license the suite for its modules and use Nanome for the review\u003C\u002Ftd>\u003C\u002Ftr>\n  \u003C\u002Ftbody>\n\u003C\u002Ftable>\n\n## What Nanome loads and connects to\n\n\u003Cdiv class=\"logo-row\" style=\"display:flex;flex-wrap:wrap;align-items:center;justify-content:center;gap:1.75rem;background:#fff;border-radius:12px;padding:1.25rem 1.5rem;margin:2rem 0\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_nanome_is_different_from_biovia_discovery_studio_image_5_269cf06656.png\" alt='Schrödinger' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_nanome_is_different_from_biovia_discovery_studio_image_6_f57b31ff45.png\" alt='OpenEye' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_nanome_is_different_from_biovia_discovery_studio_image_8_0ea60edb4d.png\" alt='CDD Vault' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_nanome_is_different_from_biovia_discovery_studio_image_9_60fa461fad.png\" alt='KNIME' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003C\u002Fdiv>\n\n\n\n![Two colleagues in a lounge area review a smooth molecular surface structure on a large wall display, one pointing at the visible binding pocket.](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_nanome_is_different_from_biovia_discovery_studio_image_2_da3db7981c.png)\n\n\n\nA structure written out of Discovery Studio opens in Nanome with no conversion step in between: `.pdb` and `.ent`, mmCIF as `.cif`, `.mmcif`, `.mcif`, or `.bcif`, then `.sdf`, `.mol`, `.mol2`, `.xyz`, `.pqr`, and `.pdbqt`. Nanome fetches by accession too, from RCSB PDB, PubChem, DrugBank, ChEMBL, UniProt, and the AlphaFold Protein Structure Database, so an ID pasted into chat turns into a shared 3D session in a couple of steps. Molecules come back out as PDB, SDF, or SMILES, one frame at a time.\n\nIf your Discovery Studio work runs molecular dynamics, the frames travel with it. Trajectories come across from your simulation engine. Load a `.gro` standalone, or attach `.xtc`, `.trr`, or `.dcd` frames to a model that's already open, matching atom counts. Frame playback caps at 2000, and surfaces are off during playback. [More on supported formats](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n\nNanome sits alongside the other software a team already licenses. The [Schrödinger](https:\u002F\u002Fwww.schrodinger.com) LiveDesign gadget hands `.mae` and `.maegz` from a LiveReport straight into Nanome. [CDD Vault](https:\u002F\u002Fwww.collaborativedrug.com), Cresset Flare, [the OpenFold Consortium](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fnanome-joins-the-openfold-consortium), and [OpenEye (Cadence)](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fview-your-openeyecadence-data-with-fresh-eyes-using-nanome-xr) connect as well. For anything homegrown, there's an open REST API, MCP servers, a Nanome Claude Code Skill, and [KNIME](https:\u002F\u002Fwww.knime.com) or Jupyter on the other end.\n\nPart of this comparison has been measured. [Kingsley et al., 2019](https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.jmgm.2019.03.010), in the Journal of Molecular Graphics and Modelling, timed Nanome's surface rendering against the then-current builds of Discovery Studio, PyMOL, Chimera, and ChimeraX, and found it several times faster, with load times close to those 2D tools even while drawing two eye views at 90+ frames per second. In the same paper, chemists rebuilt a known RIP2 kinase inhibitor from scratch inside the pocket and landed within 1.8 Å RMSD of the co-crystal structure. A Novartis GNF team sits on the author list. Every product involved has shipped years of releases since then, so read those figures as a dated measurement.\n\n## When Discovery Studio is the right pick\n\nIf your projects lean on particular Discovery Studio modules, and a licensed workstation per scientist suits how the group operates, the suite may cover you well. Validated module depth is what a broad commercial package is built to deliver, and Nanome doesn't try to match it.\n\nNanome tends to fit when several people need to inspect the same structure together, when depth perception helps read a pocket or a loop, or when a group would rather ask for an analysis in words. Because it integrates with several established suites, a team can keep the software it trusts and add shared sessions on top. The same split turns up elsewhere: [how Nanome differs from PyMOL](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fhow-nanome-is-different-from-pymol) covers a free desktop viewer, and [how Nanome differs from MOE](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fhow-nanome-is-different-from-moe) covers a suite that doubles as a Nanome integration. The [case studies](https:\u002F\u002Fnanome.ai\u002Fcase-studies) show how groups have divided the work in practice.\n\n## FAQ\n\n**What are the alternatives to Discovery Studio for molecular visualization?**\nThey run from free desktop viewers such as PyMOL, UCSF ChimeraX, and VMD to full commercial suites such as Schrödinger Maestro and MOE. Nanome's angle is a session several people join at once, native XR, and MARA taking analysis requests in ordinary words. The [comparison hub](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fhow-nanome-is-different-from-pymol) has the tool-by-tool detail.\n\n**Is Nanome a full replacement for a comp-chem suite?**\nNo. Nanome covers collaborative visualization and drives 300+ tools through MARA, and it integrates with suites like Schrödinger LiveDesign and Cresset Flare, so most teams run it beside what they already own.\n\n**Can Nanome keep our data behind our own firewall?**\nYes. Enterprise deployments run single-tenant cloud or on-prem, so structures and results stay inside your own environment.\n\n**What files and databases does Nanome support?**\nStructures import as `.pdb` and `.ent`, mmCIF (`.cif`, `.mmcif`, `.mcif`, `.bcif`), `.sdf` and `.sd`, `.mol` and `.mol2`, SMILES, `.xyz`, `.pqr`, and `.pdbqt`. Vendor and session files import as Maestro `.mae` and `.maegz`, `.moe`, and PyMOL `.pse`. MD trajectories import as `.gro` standalone or as `.xtc`, `.trr`, and `.dcd` attached to an open model, and `.dx` electrostatic maps overlay a loaded model. Export is PDB, SDF, or SMILES, single frame, so mmCIF, MAE, MOE, and PSE come in without going back out. Fetching works from RCSB PDB, PubChem, DrugBank, ChEMBL, UniProt, and the AlphaFold Protein Structure Database. [More on supported formats](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n","2026-07-15T01:23:51.150Z","2026-09-08T16:00:06.057Z","2026-09-08T16:00:05.980Z","A Discovery Studio alternative for teams who want real-time multiplayer, native XR, and a plain-English AI copilot behind their firewall.","alternatives to Discovery Studio, Discovery Studio alternative, BIOVIA Discovery Studio, Nanome, molecular visualization, drug discovery, XR molecular modeling, MARA AI copilot","how-nanome-is-different-from-biovia-discovery-studio",{"id":244,"attributes":245},93,{"title":246,"content":247,"createdAt":248,"updatedAt":249,"publishedAt":250,"date":217,"description":251,"keywords":252,"slug":253,"category":17},"How Nanome is different from MOE","MOE (Molecular Operating Environment), from [Chemical Computing Group](https:\u002F\u002Fwww.chemcomp.com), is an integrated computational chemistry suite, strong in medicinal chemistry, protein modeling, and antibody work, run through a desktop interface. Nanome is a collaborative molecular visualization and drug discovery platform that spans a browser web app, XR headsets, and Windows desktop. Inside it, [an AI copilot called MARA](https:\u002F\u002Fnanome.ai\u002Fmara) runs the computational work. If you're looking at alternatives to MOE, the two cover different parts of the problem, and Nanome often sits alongside a suite like MOE rather than replacing it. [CCG MOE is a listed integration on nanome.ai](https:\u002F\u002Fnanome.ai\u002Fintegrations), so this is a pairing as much as a choice.\n\nBoth tools read the same structure files. The difference shows up in what happens once a file is open.\n\n## What MOE does well\n\n\u003Cdiv class=\"logo-row\" style=\"display:flex;flex-wrap:wrap;align-items:center;justify-content:center;gap:1.75rem;background:#fff;border-radius:12px;padding:1.25rem 1.5rem;margin:2rem 0\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_nanome_is_different_from_moe_image_3_1d42fc1564.png\" alt='MOE' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_nanome_is_different_from_moe_image_4_3e7bf1eff3.png\" alt='GROMACS' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_nanome_is_different_from_moe_image_5_19d10757cc.png\" alt='Schrödinger Maestro' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003C\u002Fdiv>\n\n\n\nMOE is a mature, single-package environment for structure-based and ligand-based design. The draw is that the whole pipeline sits in one window: protein preparation, docking, pharmacophore work, QSAR, and antibody modeling, with SVL scripting to automate any of it.\n\nThe antibody side is the part worth naming. MOE has a long track record across medicinal chemistry and biologics, and a lot of published work runs through it. A group fluent in SVL, with its workflows already built around the suite, has a real reason to stay put.\n\n## Where Nanome is different\n\n![Two colleagues wearing ultra-thin VR headsets examine the same ribbon-cartoon protein structure floating between them](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_nanome_is_different_from_moe_image_1_v4s_45bb534f3a.png)\n\n\n\nNanome covers ground a desktop suite doesn't aim at.\n\n**Real-time multiplayer.** Several people stand inside the same structure at once and point at the same atoms. Everyone holds the model, so a colleague can swing the ligand around while you watch the pocket from the far side.\n\n**Native XR.** Nanome [runs in immersive 3D on Apple Vision Pro, Meta Quest, HTC Vive Focus 3, and Pico Neo, and on Windows desktop or in a browser for anyone without a headset](https:\u002F\u002Fnanome.ai\u002Fsetup). You walk around a binding pocket at hand scale instead of rotating it on a flat monitor.\n\n**An AI copilot (MARA).** Ask in plain English and MARA runs the job across [300+ integrated scientific tools](https:\u002F\u002Fnanome.ai\u002Fintegrations): docking with Smina and DiffDock-L, electrostatics through APBS, ADMET and toxicity prediction, sequence-to-structure folding with AlphaFold 3 and Boltz-2, and de novo binder design with RFdiffusion3 (beta). Each run reports the tool that did the work, the inputs it took, and what it returned, so the result can be checked rather than taken on faith.\n\n**Antibody tooling, specifically.** Since antibody modeling is MOE's strongest suit, that comparison deserves detail. [ProteinMPNN designs new sequences for a given backbone, and ANARCI numbers a variable domain and defines its CDR loops](https:\u002F\u002Fdocs.nanome.ai\u002Fmara\u002Ffeatures) under the Kabat, Chothia, IMGT, and AHo schemes. One project reviewed this way is [antibody prophylaxis for Lyme disease](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fantibody-prophylaxis-for-lyme-disease), where the antibody is meant to block infection before it takes hold and the binding interface was examined in immersive 3D.\n\n## Side by side\n\n\u003Ctable class=\"table\">\n  \u003Cthead>\n    \u003Ctr>\u003Cth>Package\u003C\u002Fth>\u003Cth>Where it's strong\u003C\u002Fth>\u003Cth>How the pair works\u003C\u002Fth>\u003C\u002Ftr>\n  \u003C\u002Fthead>\n  \u003Ctbody>\n    \u003Ctr>\u003Ctd>MOE\u003C\u002Ftd>\u003Ctd>One desktop suite for medicinal chemistry: protein prep, docking, pharmacophore and QSAR work, antibody modeling, all scriptable in SVL\u003C\u002Ftd>\u003Ctd>Nanome imports .moe files, so a structure prepared in MOE opens in a shared session without a conversion step\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Nanome\u003C\u002Ftd>\u003Ctd>Multi-user visualization in a browser and in headsets, with MARA reaching 300+ tools including ProteinMPNN and ANARCI\u003C\u002Ftd>\u003Ctd>MARA covers docking, folding, electrostatics, and antibody numbering from a plain-English request\u003C\u002Ftd>\u003C\u002Ftr>\n  \u003C\u002Ftbody>\n\u003C\u002Ftable>\n\n## When MOE is the right pick\n\n![Two colleagues wearing ultra-thin VR headsets examine the same space-filling protein model floating between them](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_nanome_is_different_from_moe_image_2_v4s_ec5aaa12c2.png)\n\n\n\nA team that wants one desktop suite with a deep, established medicinal chemistry and antibody pipeline, staffed by people fluent in SVL, is well served by MOE. It's respected for good reasons.\n\nNanome fits when the work is collaborative, when seeing a structure at hand scale changes what someone notices about a pocket, or when you'd rather ask for docking, folding, and antibody annotation than write the script. Plenty of groups keep a suite like MOE for the heavy modeling and bring the structures into Nanome for the shared review.\n\nIf that modeling ends in molecular dynamics, the trajectory can come along. Nanome loads and plays back MD trajectories. A `.gro` file loads on its own; `.xtc`, `.trr`, and `.dcd` attach to a model already in the workspace and have to match its atom count. Frame-trajectory playback carries a 2000-frame cap, and surfaces switch off while a trajectory plays. [More on supported formats](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n\nTwo neighboring comparisons cover the rest of this category: the [PyMOL post](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fhow-nanome-is-different-from-pymol) for the desktop viewer side, and the [BIOVIA Discovery Studio post](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fhow-nanome-is-different-from-biovia-discovery-studio) for the other broad commercial package.\n\nStructures arrive from PubChem, DrugBank, ChEMBL, UniProt, and the RCSB PDB without a separate download, and an open REST API plus MCP servers let a script or an agent drive the workspace directly. Nanome's [case studies](https:\u002F\u002Fnanome.ai\u002Fcase-studies) collect the customer project write-ups.\n\n## FAQ\n\n**What are the alternatives to MOE?**\nFull desktop comp-chem suites like [Schrödinger Maestro](https:\u002F\u002Fwww.schrodinger.com\u002Fplatform\u002Fproducts\u002Fmaestro\u002F) and [BIOVIA Discovery Studio](https:\u002F\u002Fwww.3ds.com\u002Fproducts\u002Fbiovia\u002Fdiscovery-studio) cover similar ground to MOE. Nanome comes at the same structures from a different angle, with multi-user visualization on the web and in XR and MARA driving the computational tools, so teams more often pair it with a suite than swap one out. Files move either way: Nanome imports Maestro `.mae` and `.maegz` (the LiveDesign ingestion format) alongside PDB and mmCIF coordinates, SDF and MOL2 ligands, plus XYZ and PQR.\n\n**What file formats does Nanome support?**\nStructures: PDB (`.pdb`, `.ent`), mmCIF (`.cif`, `.mmcif`, `.bcif`), SDF (`.sdf`, `.sd`), MOL and MOL2, SMILES, XYZ, PQR, and PDBQT. Vendor and session files are import-only: Maestro `.mae` and `.maegz`, MOE `.moe`, and PyMOL `.pse`. MD trajectories: a `.gro` file stands alone, while `.xtc`, `.trr`, and `.dcd` attach to a model that is already loaded and must match its atom count. Electrostatic maps attach as `.dx` overlays. Export is PDB, SDF, or SMILES, one frame at a time. [More on supported formats](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n\n**Is Nanome a Molecular Operating Environment alternative for antibody work?**\nIn part. ProteinMPNN designs sequences, ANARCI numbers variable domains and defines CDR loops, and the Lyme prophylaxis project shows what the shared 3D review adds on top. MOE remains a strong dedicated choice for antibody modeling, which is why a lot of groups run both.\n\n**Can I use MOE and Nanome together?**\nYes. CCG MOE is a listed integration on nanome.ai. Nanome imports `.moe` files directly, so a structure prepared in MOE opens in a shared XR session, and MARA can then dock it, fold a partner, or run electrostatics on it. An open REST API and MCP servers cover the scripted path.\n\n**Does Nanome need a headset?**\nNo. It runs in a browser web app and on Windows desktop. XR is there when immersive 3D helps, on Meta Quest, Apple Vision Pro, HTC Vive Focus 3, or Pico Neo.\n","2026-07-15T01:23:51.301Z","2026-09-08T16:00:07.790Z","2026-09-08T16:00:07.741Z","A fair MOE alternative guide: where the Molecular Operating Environment fits, and where Nanome adds multiplayer XR and the MARA AI copilot.","alternatives to MOE, Molecular Operating Environment alternative, MOE alternative, Nanome, MARA, molecular modeling, antibody modeling","how-nanome-is-different-from-moe",{"id":255,"attributes":256},75,{"title":257,"content":258,"createdAt":259,"updatedAt":260,"publishedAt":261,"date":262,"description":263,"keywords":264,"slug":265,"category":17},"An AI copilot for drug discovery workflows","An AI copilot for drug discovery is an assistant that takes a plain-English request and runs the actual scientific tools behind it, then shows you exactly what it did. [Nanome's copilot is called MARA](https:\u002F\u002Fnanome.ai\u002Fmara). It reaches [300+ integrated scientific tools](https:\u002F\u002Fnanome.ai\u002Fintegrations), runs them from ordinary sentences, and reports which tools fired, what inputs they got, and what came back.\n\nThat last part matters. A lot of assistants can talk about docking. MARA runs the docking, then hands you the poses.\n\n## What an AI copilot actually does\n\n![A researcher wearing an ultra-thin VR headset studies a ribbon-cartoon protein structure held close to the chest](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fan_ai_copilot_for_drug_discovery_workflows_image_1_v4s_c826eb2bf4.png)\n\n\n\nThe job is orchestration. You describe the goal, and the copilot picks the right tools, chains them in order, and returns real molecular results you can inspect in 3D.\n\nThe built-in library covers what a medicinal chemist or structural biologist works with day to day:\n\n- **Docking.** Smina and DiffDock-L place a ligand in a pocket and score how well it sits.\n- **Co-folding.** A protein and its ligand fold together, which predicts the bound complex directly.\n- **Structure prediction.** [Boltz-2](https:\u002F\u002Fgithub.com\u002Fjwohlwend\u002Fboltz), [AlphaFold 3](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41586-024-07487-w), OpenFold3, Chai-1, ESMFold and Protenix are separate calls, so running two and comparing them is a normal afternoon.\n- **De novo binder design.** RFdiffusion3 (beta) generates candidate binders, and ProteinMPNN writes sequences onto a backbone.\n- **Antibody annotation.** ANARCI numbers variable domains under Kabat, Chothia, IMGT or AHo and marks out the CDR loops.\n- **Electrostatics.** APBS computes charge and potential surfaces, with PDB2PQR ahead of it for protonation.\n- **ADMET and toxicity.** Absorption, distribution, metabolism and excretion estimates, plus eToxPred, before a molecule gets committed to synthesis.\n- **Cheminformatics.** Property calculations and the routine bookkeeping around a series.\n\nYou ask in English. MARA translates that into a sequence of tool calls and runs them.\n\nThat library is a starting point. Groups write MARA tools of their own, wrapping an in-house method or a model the group trained, and publish them so anyone in the org can call the thing by name. Stringing those calls into a longer autonomous run has its own considerations, covered in [agentic AI for computational chemistry](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fagentic-ai-for-computational-chemistry).\n\n## A visible record of every run\n\n![A flat vector diagram showing three scientific tools each connected to their inputs and outputs in a transparent, auditable chain.](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fan_ai_copilot_for_drug_discovery_workflows_image_2_89e053fe60.png)\n\n\n\nA copilot is worth pointing at real science when its work is auditable. MARA names the tool it called, the inputs it passed, and the artifact it produced, and every one of those is something a scientist can open.\n\nWhen it reports a docking score, the pose, the parameters and the source structure come with it. When it predicts a fold, the report says whether Boltz-2 or AlphaFold 3 produced the answer. Both are results a chemist can pull apart before acting on them.\n\nThat review step is the whole argument for a copilot over a chatbot. Insilico Medicine generated 10 novel SARS-CoV-2 protease inhibitors with AI, then brought them into Nanome in VR for the medicinal chemistry review. The work was co-authored and posted to ChemRxiv. Their CEO put the reasoning plainly:\n\n> \"While AI can come up with novel and diverse drug-like molecules, it is important for medicinal chemists to look at these molecules closely before placing a billion-dollar, life-or-death wager. VR enables medicinal chemists to do this.\"\n>\n> Alex Zhavoronkov, CEO, Insilico Medicine\n\n## Deployment behind your firewall\n\nDrug discovery data is sensitive, so MARA and Nanome's enterprise deployments run [behind your firewall](https:\u002F\u002Fnanome.ai\u002Fmara-enterprise), as a single-tenant cloud instance or entirely on-prem. Structures and sequences stay in the environment they started in.\n\nThat is what lets teams at Genentech and [Novartis](https:\u002F\u002Fnanome.ai\u002Fcase-studies\u002Fnovartis) point a copilot at proprietary chemistry without shipping it to a shared service. The peer-reviewed record sits in Nanome's [publications](https:\u002F\u002Fnanome.ai\u002Fpublications).\n\n## 300+ tools, an open API, and MCP servers\n\nMARA connects to 300+ scientific tools across 26 categories. Nanome also ships an open REST API plus MCP servers, so the copilot can be driven from the pipeline you already run, or drive it.\n\nStructures come in from RCSB PDB, PubChem and DrugBank without a download step, and Nanome reads PDB, SDF, mmCIF, MOL\u002FMOL2, PQR, XYZ and SMILES. [More on supported formats](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n\nIt sits next to suites teams already own: [CDD Vault](https:\u002F\u002Fwww.collaborativedrug.com) and [KNIME](https:\u002F\u002Fwww.knime.com) on the data side, [Schrödinger LiveDesign](https:\u002F\u002Fnanome.ai\u002Fintegrations), Cresset Flare and [OpenEye](https:\u002F\u002Fwww.eyesopen.com) (Cadence) on the modeling side, plus [OpenFold](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fnanome-joins-the-openfold-consortium) and Jupyter. A Nanome Claude Code Skill is available too, so an outside agent stack can call Nanome tools directly. There's a fuller walkthrough in [MCP servers for cheminformatics and drug discovery](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fmcp-servers-for-cheminformatics-and-drug-discovery).\n\n## How MARA differs from a general chat assistant\n\n\u003Ctable class=\"table\">\n  \u003Cthead>\n    \u003Ctr>\u003Cth>Capability\u003C\u002Fth>\u003Cth>Generic chat assistant\u003C\u002Fth>\u003Cth>Nanome MARA\u003C\u002Fth>\u003C\u002Ftr>\n  \u003C\u002Fthead>\n  \u003Ctbody>\n    \u003Ctr>\u003Ctd>Answer questions about docking or folding\u003C\u002Ftd>\u003Ctd>Yes\u003C\u002Ftd>\u003Ctd>Yes\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Actually run docking, co-folding, ADMET\u003C\u002Ftd>\u003Ctd>No\u003C\u002Ftd>\u003Ctd>Yes, 300+ tools\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Name the tool it called and the inputs it used\u003C\u002Ftd>\u003Ctd>No\u003C\u002Ftd>\u003Ctd>Yes, full trace\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Return inspectable 3D molecular results\u003C\u002Ftd>\u003Ctd>No\u003C\u002Ftd>\u003Ctd>Yes, in the web app and XR\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Run behind your firewall on your data\u003C\u002Ftd>\u003Ctd>Rarely\u003C\u002Ftd>\u003Ctd>Yes, single-tenant or on-prem\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Plug into your pipeline via API and MCP\u003C\u002Ftd>\u003Ctd>Limited\u003C\u002Ftd>\u003Ctd>Yes, open REST API plus MCP servers\u003C\u002Ftd>\u003C\u002Ftr>\n  \u003C\u002Ftbody>\n\u003C\u002Ftable>\n\nA general assistant can describe a method well. MARA runs it and leaves you the files.\n\n## Where a different tool fits better\n\n\u003Cdiv class=\"logo-row\" style=\"display:flex;flex-wrap:wrap;align-items:center;justify-content:center;gap:1.75rem;background:#fff;border-radius:12px;padding:1.25rem 1.5rem;margin:2rem 0\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fan_ai_copilot_for_drug_discovery_workflows_image_4_3d89830a22.png\" alt='PyMOL' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fan_ai_copilot_for_drug_discovery_workflows_image_6_3aa56cfacb.png\" alt='Schrödinger Maestro' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fan_ai_copilot_for_drug_discovery_workflows_image_7_15bb662f5f.png\" alt='MOE' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fan_ai_copilot_for_drug_discovery_workflows_image_9_08f64122bd.png\" alt='OpenEye' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003C\u002Fdiv>\n\n\n\nFor a scriptable desktop viewer under one person's control, [PyMOL](https:\u002F\u002Fpymol.org) and [UCSF ChimeraX](https:\u002F\u002Fwww.rbvi.ucsf.edu\u002Fchimerax\u002F) are excellent and heavily documented. Nanome opens PyMOL `.pse` session files for viewing, and it parses the standard structure files ChimeraX handles.\n\nWhen a specialized workflow lives inside a full comp-chem suite like [Schrödinger](https:\u002F\u002Fwww.schrodinger.com) [Maestro](https:\u002F\u002Fwww.schrodinger.com\u002Fplatform\u002Fproducts\u002Fmaestro\u002F) or [MOE](https:\u002F\u002Fwww.chemcomp.com), that suite stays the right home for it. Nanome imports Maestro `.mae` and `.maegz` (the format LiveDesign ingests) plus `.moe`, so prepared work opens in a shared 3D session without a rebuild. [More on supported formats](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats). Several of those suites are Nanome integrations, so most groups run both.\n\nA copilot pays off when the work crosses many tools, when the record of what ran has to be legible to a colleague who wasn't there, and when the people judging the result want to stand around the same molecule in a browser or a headset. The customer proof for all of that sits in the [case studies](https:\u002F\u002Fnanome.ai\u002Fcase-studies).\n\n## FAQ\n\n**What is an AI copilot for drug discovery workflows?**\nAn assistant that turns a plain-English request into real tool runs across docking, folding, ADMET and design work, then shows what came back. Nanome's copilot, MARA, orchestrates 300+ scientific tools spread across 26 categories.\n\n**What scientific tools can MARA run?**\nStructure prediction and co-folding (AlphaFold 3, Boltz-2, OpenFold3, Chai-1, ESMFold, Protenix), docking (Smina, DiffDock-L), de novo design (RFdiffusion3, ProteinMPNN), antibody numbering and CDR definition (ANARCI), electrostatics (APBS, PDB2PQR), toxicity and ADMET estimates (eToxPred, ToxinPred), and cheminformatics property work.\n\n**Can these AI-powered molecular analysis tools run on private data?**\nYes. Enterprise deployments of Nanome and MARA sit inside your own network, either as a single-tenant cloud instance or fully on-premises, so nothing proprietary leaves the building.\n\n**How does MARA fit into an existing pipeline?**\nThrough the REST API, the MCP servers, and direct integrations with CDD Vault, KNIME, Jupyter, Schrödinger LiveDesign, Cresset Flare and OpenEye (Cadence). A Nanome Claude Code Skill is available for teams driving Nanome from their own agents.\n","2026-07-15T01:23:49.403Z","2026-09-03T16:00:26.033Z","2026-09-03T16:00:25.781Z","2026-09-03","MARA is Nanome's AI copilot for drug discovery workflows: plain-English orchestration of docking, co-folding, ADMET, and structure prediction.","AI copilot for drug discovery workflows, AI-powered molecular analysis tools, MARA, Nanome, molecular docking AI, co-folding, ADMET prediction, structure prediction, MCP servers","an-ai-copilot-for-drug-discovery-workflows",{"id":267,"attributes":268},87,{"title":269,"content":270,"createdAt":271,"updatedAt":272,"publishedAt":273,"date":262,"description":274,"keywords":275,"slug":276,"category":17},"MCP servers for cheminformatics and drug discovery","Nanome offers MCP servers and a Claude Code Skill that let an AI model drive real cheminformatics and drug discovery tools: docking, structure prediction, and analysis. MCP (Model Context Protocol) is the open standard that connects a language model to outside tools and data, so instead of an LLM guessing at chemistry, it calls the actual programs through Nanome's AI copilot, MARA, which reaches [more than 300 integrated scientific tools](https:\u002F\u002Fnanome.ai\u002Fintegrations). The servers [install inside your own network](https:\u002F\u002Fnanome.ai\u002Fmara-enterprise), as a single tenant in the cloud or on hardware your group owns, and they take an internal or custom LLM as the model behind them.\n\n## What an MCP server does for chemistry\n\n![A plain-language request passes through an MCP gateway and triggers three distinct scientific tools on the other side.](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fmcp_servers_for_cheminformatics_and_drug_discovery_image_1_db521186a2.png)\n\n\n\nMCP (Model Context Protocol) is a shared way for an AI model to talk to external tools. You stand up an MCP server that lists a set of functions, and any MCP-aware client (Claude Code, other agents, your own app) can call them.\n\nA language model has no docking engine inside it. It can produce a fluent paragraph about a binding pose having run no calculation at all. An MCP server puts a working engine on the other end of the call: the model chooses the function and reads what comes back, and the chemistry belongs to the program that ran it.\n\nThat is what Nanome ships. The MCP server hands MARA's tools to whatever model you point at it, so an AI can drive a docking run, kick off a structure prediction, or fetch a compound and open it up, all from plain-English requests.\n\n## What the Nanome MCP tools expose\n\nThrough MARA, the server puts these categories within reach of a connected model:\n\n- **Docking.** [Pose prediction and scoring for a ligand against a target pocket](https:\u002F\u002Fdocs.nanome.ai\u002Fmara\u002Ffeatures), with Smina and DiffDock-L behind it.\n- **Structure prediction and co-folding.** [Boltz-2 and AlphaFold 3 fold a sequence](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fsetting-up-boltz-2-configuration-files-and-analysis-with-nanome-ai), and either one will fold a protein together with its ligand to predict the bound complex.\n- **Electrostatics.** APBS solves the surface charge on a protein, with PDB2PQR ahead of it to sort out protonation.\n- **ADMET prediction.** Absorption, distribution, metabolism, excretion, and toxicity estimates on a candidate compound.\n- **Antibody annotation and binder design.** [ANARCI numbers variable domains and marks the CDR loops, ProteinMPNN writes sequences onto a backbone, and RFdiffusion3 (beta) generates de novo binders](https:\u002F\u002Fnanome.ai\u002Fagents).\n- **Cheminformatics.** Descriptors, similarity, filters, R-group work, and matched molecular pairs.\n- **Data pulls.** A model can fetch a structure from RCSB PDB, a compound from PubChem, or a record from DrugBank, and it opens in the session ready to inspect.\n\nEvery call leaves a trace. MARA names the tool it fired, the arguments it passed, and the artifact that came back, so a result delivered over MCP can be checked the same way as one produced at a terminal. The same library seen from inside the product, rather than through the protocol, gets walked through in [an AI copilot for drug discovery workflows](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fan-ai-copilot-for-drug-discovery-workflows).\n\nGroups add tools of their own. An in-house scoring script, or a model a team trained on its own series, gets wrapped as a MARA tool with a name and a signature, and the server then lists it beside the standard ones for any connected model to call.\n\n## MCP server or Claude Code Skill\n\nNanome ships 2 ways to connect a model, and they answer slightly different questions.\n\n\u003Ctable class=\"table\">\n  \u003Cthead>\n    \u003Ctr>\u003Cth>Option\u003C\u002Fth>\u003Cth>What it is\u003C\u002Fth>\u003Cth>Best fit\u003C\u002Fth>\u003C\u002Ftr>\n  \u003C\u002Fthead>\n  \u003Ctbody>\n    \u003Ctr>\u003Ctd>Nanome MCP server\u003C\u002Ftd>\u003Ctd>An MCP endpoint that lists MARA's tools to any MCP-aware client\u003C\u002Ftd>\u003Ctd>A model, an agent, or your own application has to call docking, folding, and analysis programmatically\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Nanome Claude Code Skill\u003C\u002Ftd>\u003Ctd>A packaged Skill that teaches Claude Code how to drive Nanome and MARA\u003C\u002Ftd>\u003Ctd>Claude Code is already the working environment, and chemistry steps belong inside a coding or analysis session\u003C\u002Ftd>\u003C\u002Ftr>\n  \u003C\u002Ftbody>\n\u003C\u002Ftable>\n\nThe MCP server is the general connector, usable from anything that speaks the protocol. The Skill is the shorter path when Claude Code is where the work already happens. Chaining several calls into one longer autonomous run carries its own tradeoffs, covered in [agentic AI for computational chemistry](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fagentic-ai-for-computational-chemistry).\n\n## Your network, your model\n\n2 constraints tend to govern this work once the compounds are proprietary and the target is unpublished.\n\nDeployment is the first. Nanome, MARA, and the MCP servers install inside your own network, as a single tenant in the cloud or on hardware your group owns. A docking run never has to cross the boundary to finish.\n\nThe reasoning layer is the second. MCP is model-agnostic, so the same Nanome tools answer to a hosted commercial model or to a fine-tuned one on your own GPUs. The server takes whichever model a security review has already cleared.\n\nTogether those two keep an AI-driven run inside the same perimeter the rest of the program already operates in.\n\n## When a direct tool call is simpler\n\n\u003Cdiv class=\"logo-row\" style=\"display:flex;flex-wrap:wrap;align-items:center;justify-content:center;gap:1.75rem;background:#fff;border-radius:12px;padding:1.25rem 1.5rem;margin:2rem 0\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fmcp_servers_for_cheminformatics_and_drug_discovery_image_5_baebdd1ecd.png\" alt='CDD Vault' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fmcp_servers_for_cheminformatics_and_drug_discovery_image_6_4d4c1ec5ed.png\" alt='KNIME' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fmcp_servers_for_cheminformatics_and_drug_discovery_image_7_031da381cb.png\" alt='Jupyter' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fmcp_servers_for_cheminformatics_and_drug_discovery_image_8_463faa8451.png\" alt='GROMACS' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003C\u002Fdiv>\n\n\n\nA fixed pipeline that fires the same docking job every night is served fine by a scripted call to the program underneath. Routing it through a model adds a decision step the job never needed. MCP pays off on exploratory work, where what runs next depends on what the last run returned.\n\nNanome sits next to the software a group already owns. It connects to [Schrödinger LiveDesign](https:\u002F\u002Fnanome.ai\u002Fintegrations), [CDD Vault](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fcollaborative-drug-discovery-and-nanome-partnership-announcement), [KNIME](https:\u002F\u002Fwww.knime.com), and [Jupyter](https:\u002F\u002Fjupyter.org), so the common arrangement keeps all of that and adds the MCP server for the work where a model picks the next step.\n\nOn the Schrödinger side, `.mae` and `.maegz` files open in Nanome directly, and that is the same format LiveDesign hands structures over in, so prepared work arrives without a rebuild. [More on supported formats](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n\n## Proven on real molecules\n\n![Two colleagues review a ribbon-rendered protein structure together on a large display in a modern lounge setting.](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fmcp_servers_for_cheminformatics_and_drug_discovery_image_3_b584d7a212.png)\n\n\n\nInsilico Medicine generated 10 novel SARS-CoV-2 protease inhibitors with AI, then brought them into Nanome for the medicinal chemistry review before anything went to synthesis. The work was co-authored and posted to ChemRxiv. Alex Zhavoronkov, the company's CEO, gave the reason for that step: \"it is important for medicinal chemists to look at these molecules closely before placing a billion-dollar, life-or-death wager.\"\n\nThat review is the argument for wiring a model to real tools. A generated molecule still has to be judged by a chemist, and a call that returns a real structure puts the thing itself in front of them. What other groups have done with the platform is collected in the [case studies](https:\u002F\u002Fnanome.ai\u002Fcase-studies).\n\n## FAQ\n\n**Is there an MCP server for cheminformatics?**\nYes. Nanome runs one, and it exposes cheminformatics and drug discovery tools through MARA. A connected model can calculate descriptors, run similarity searches and filters, dock a ligand, predict a structure, and work through the rest of a library spanning more than 300 integrated tools across 26 categories.\n\n**How do I connect an LLM to chemistry tools?**\nAny MCP-aware client can point at the Nanome MCP server, and teams already working in Claude Code can install the Nanome Claude Code Skill instead. Either way the model calls docking, folding, electrostatics, and ADMET as functions, and the numbers come back from the programs themselves.\n\n**Can the MCP server run on private data?**\nYes. Nanome, MARA, and the MCP servers install inside your own network, as a single tenant in the cloud or on hardware you own, and an internal or custom LLM can drive them, so proprietary structures stay where they started.\n\n**What can an AI actually do through the Nanome MCP server?**\nDock a ligand (Smina, DiffDock-L), predict a structure or co-fold a complex ([Boltz-2](https:\u002F\u002Fgithub.com\u002Fjwohlwend\u002Fboltz), [AlphaFold 3](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41586-024-07487-w)), solve electrostatics with APBS, estimate ADMET and toxicity, number antibody variable domains with ANARCI, design sequences onto a backbone with ProteinMPNN, generate binders with RFdiffusion3, and run cheminformatics calculations, each with a record of the call and what it returned.\n","2026-07-15T01:23:50.592Z","2026-09-03T16:00:28.589Z","2026-09-03T16:00:28.330Z","MCP server for cheminformatics and drug discovery: Nanome exposes docking, structure prediction, and analysis to an LLM behind your firewall.","MCP server cheminformatics, MCP server drug discovery, connect an LLM to chemistry tools, Model Context Protocol, Claude Code Skill, Nanome, MARA, docking, structure prediction","mcp-servers-for-cheminformatics-and-drug-discovery",{"id":278,"attributes":279},74,{"title":280,"content":281,"createdAt":282,"updatedAt":283,"publishedAt":284,"date":262,"description":285,"keywords":286,"slug":287,"category":17},"Agentic AI for computational chemistry","Agentic AI for computational chemistry is an AI that plans a scientific task, picks the right tools, and runs them on real structures, then hands you the results. [Nanome's MARA](https:\u002F\u002Fnanome.ai\u002Fmara) is one of these. It's an AI copilot inside Nanome's molecular visualization and drug discovery platform, and it orchestrates [300+ integrated scientific tools](https:\u002F\u002Fnanome.ai\u002Fintegrations) from plain-English requests, across [the web app and XR headsets](https:\u002F\u002Fnanome.ai\u002Fsetup), behind your firewall.\n\nAsk a language model about docking and you get prose. Ask MARA, and a docking job runs against your protein and your ligand, then the pose comes back in 3D.\n\n## What \"agentic\" means here\n\n![A researcher wearing an ultra-thin VR headset studies a solid protein surface model with a small ligand visible in its binding pocket held close to the chest](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fagentic_ai_for_computational_chemistry_image_1_v4_d9a4b5a5c4.png)\n\n\n\nTwo separable jobs sit behind any useful answer. Something has to decide what to run, and something has to run it. A language model handles the first well and has no way to do the second on its own.\n\nAgentic AI wires the two together. You describe the goal (\"dock this ligand into the ATP pocket\", \"design a binder for this epitope\"), and MARA chooses which of the 300+ tools to call, in what order, with what inputs. It fires [docking engines, folding models and electrostatics solvers](https:\u002F\u002Fnanome.ai\u002Fagents) along with the cheminformatics routines around them, then hands back structures you can turn over in 3D. The day-to-day shape of that job gets walked through in [an AI copilot for drug discovery workflows](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fan-ai-copilot-for-drug-discovery-workflows).\n\nEvery step stays on the record. MARA names each tool it called, the inputs it passed, and what came back, so a finished dock arrives with its scoring function and its poses, and every line of the summary has a run behind it you can open.\n\n## What MARA can run\n\n\u003Ctable class=\"table\">\n  \u003Cthead>\n    \u003Ctr>\u003Cth>Task\u003C\u002Fth>\u003Cth>What the agent does\u003C\u002Fth>\u003Cth>Example tools\u003C\u002Fth>\u003C\u002Ftr>\n  \u003C\u002Fthead>\n  \u003Ctbody>\n    \u003Ctr>\u003Ctd>Docking\u003C\u002Ftd>\u003Ctd>Places a ligand in a pocket and scores the poses\u003C\u002Ftd>\u003Ctd>Smina, DiffDock-L, \u003Ca href=\"https:\u002F\u002Fvina.scripps.edu\">Vina\u003C\u002Fa>-style engines\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Co-folding\u003C\u002Ftd>\u003Ctd>Folds a protein and a ligand together to predict the bound complex\u003C\u002Ftd>\u003Ctd>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fjwohlwend\u002Fboltz\">Boltz-2\u003C\u002Fa>, \u003Ca href=\"https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41586-024-07487-w\">AlphaFold 3\u003C\u002Fa>, Protenix\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>De novo binder design\u003C\u002Ftd>\u003Ctd>Generates backbones and sequences for a target pocket or epitope\u003C\u002Ftd>\u003Ctd>RFdiffusion3 (beta), ProteinMPNN\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Electrostatics\u003C\u002Ftd>\u003Ctd>Solves and maps surface charge\u003C\u002Ftd>\u003Ctd>APBS\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>ADMET\u003C\u002Ftd>\u003Ctd>Estimates absorption, distribution, metabolism, excretion and toxicity\u003C\u002Ftd>\u003Ctd>ADMET models, eToxPred\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Antibody annotation\u003C\u002Ftd>\u003Ctd>Numbers variable domains and defines CDR loops with ANARCI, then designs new sequences with ProteinMPNN\u003C\u002Ftd>\u003Ctd>ANARCI, ProteinMPNN\u003C\u002Ftd>\u003C\u002Ftr>\n  \u003C\u002Ftbody>\n\u003C\u002Ftable>\n\nThe built-in library spans 300+ tools across 26 categories, and a team's own code can join it. A script or a model a group already trusts becomes a MARA tool, and from then on anyone in the org calls it by asking for it in a sentence.\n\nTwo runs show the shape of this.\n\n**A single job.** You load a protein from RCSB PDB and a ligand from PubChem, then ask for a dock. MARA finds the pockets, sets a box, runs the job, and returns ranked poses with scores. Nanome puts the top pose in front of you, in the web app or in a headset, at whatever size you want to look at it.\n\n**A chained job.** You point at an epitope and ask for new binders. RFdiffusion3 generates backbones, ProteinMPNN designs sequences onto them, and the candidates go to a folding model to see which ones hold their shape. Three tools, one request, and every intermediate structure stays in the workspace.\n\n## Behind your firewall\n\n![A flat vector diagram of a building outline containing a protein shape and a pipeline icon, illustrating that data and tool runs stay inside a secure network boundary.](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fagentic_ai_for_computational_chemistry_image_2_31a84a3395.png)\n\n\n\nNanome enterprise deployments sit inside your own network, [as a single tenant in the cloud or on hardware you run](https:\u002F\u002Fnanome.ai\u002Fmara-enterprise). Structures, prompts and tool outputs stay there. On an unpublished target that constraint decides whether an agent is usable at all, so the pipeline runs where the data already lives.\n\nNanome also ships MCP servers, a REST API and a Claude Code Skill, so the tool an analyst calls in a sentence can be called from a script too. That wiring has its own write-up in [MCP servers for cheminformatics and drug discovery](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fmcp-servers-for-cheminformatics-and-drug-discovery).\n\n## Where other tools fit\n\n\u003Cdiv class=\"logo-row\" style=\"display:flex;flex-wrap:wrap;align-items:center;justify-content:center;gap:1.75rem;background:#fff;border-radius:12px;padding:1.25rem 1.5rem;margin:2rem 0\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fagentic_ai_for_computational_chemistry_image_4_bb3229892e.png\" alt='Schrödinger Maestro' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fagentic_ai_for_computational_chemistry_image_5_1c25585668.png\" alt='PyMOL' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fagentic_ai_for_computational_chemistry_image_7_702d77a3cd.png\" alt='VMD' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fagentic_ai_for_computational_chemistry_image_8_2c09039bec.png\" alt='CDD Vault' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fagentic_ai_for_computational_chemistry_image_9_76a4c0faf4.png\" alt='GROMACS' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003C\u002Fdiv>\n\n\n\n![Two colleagues in a lounge area review a protein ribbon structure on a large display, one pointing at the fold with a stylus.](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fagentic_ai_for_computational_chemistry_image_3_1b33683e02.png)\n\n\n\nNanome sits alongside the comp-chem suites a group already runs. Several of them are direct integrations.\n\n\u003Ctable class=\"table\">\n  \u003Cthead>\n    \u003Ctr>\u003Cth>Tool\u003C\u002Fth>\u003Cth>Strength\u003C\u002Fth>\u003Cth>How Nanome connects\u003C\u002Fth>\u003C\u002Ftr>\n  \u003C\u002Fthead>\n  \u003Ctbody>\n    \u003Ctr>\n      \u003Ctd>\u003Ca href=\"https:\u002F\u002Fwww.schrodinger.com\u002Fplatform\u002Fproducts\u002Fmaestro\u002F\">Schrödinger Maestro\u003C\u002Fa> \u002F LiveDesign\u003C\u002Ftd>\n      \u003Ctd>Physics-based modeling at depth, plus enterprise data\u003C\u002Ftd>\n      \u003Ctd>Nanome imports Maestro \u003Ccode>.mae\u003C\u002Fcode> \u002F \u003Ccode>.maegz\u003C\u002Fcode> files (also the LiveDesign ingestion format) and integrates with \u003Ca href=\"https:\u002F\u002Fnanome.ai\u002Fintegrations\">LiveDesign\u003C\u002Fa>. MARA adds plain-English orchestration on top, and the review happens in 3D. \u003Ca href=\"https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats\">More on supported formats\u003C\u002Fa>\u003C\u002Ftd>\n    \u003C\u002Ftr>\n    \u003Ctr>\n      \u003Ctd>\u003Ca href=\"https:\u002F\u002Fpymol.org\">PyMOL\u003C\u002Fa>, \u003Ca href=\"https:\u002F\u002Fwww.rbvi.ucsf.edu\u002Fchimerax\u002F\">ChimeraX\u003C\u002Fa>, \u003Ca href=\"https:\u002F\u002Fwww.ks.uiuc.edu\u002FResearch\u002Fvmd\u002F\">VMD\u003C\u002Fa>\u003C\u002Ftd>\n      \u003Ctd>Scriptable desktop visualization\u003C\u002Ftd>\n      \u003Ctd>PyMOL \u003Ccode>.pse\u003C\u002Fcode> sessions import for viewing. PDB, mmCIF, SDF, MOL2, PQR and XYZ load as structures you can edit. Molecules come back out as PDB, SDF or SMILES, one frame at a time. Multiplayer sessions and XR are the part Nanome adds. \u003Ca href=\"https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats\">More on supported formats\u003C\u002Fa>\u003C\u002Ftd>\n    \u003C\u002Ftr>\n    \u003Ctr>\n      \u003Ctd>\u003Ca href=\"https:\u002F\u002Fwww.collaborativedrug.com\">CDD Vault\u003C\u002Fa>\u003C\u002Ftd>\n      \u003Ctd>Registration and assay data management\u003C\u002Ftd>\n      \u003Ctd>Nanome connects to \u003Ca href=\"https:\u002F\u002Fnanome.ai\u002Fblog\u002Fcollaborative-drug-discovery-and-nanome-partnership-announcement\">CDD Vault\u003C\u002Fa>, so an agent run lands next to the assay record it belongs with\u003C\u002Ftd>\n    \u003C\u002Ftr>\n  \u003C\u002Ftbody>\n\u003C\u002Ftable>\n\nThe payoff lands on runs that cross several of those tools at once. Each step stays visible, and the people who have to agree on the result can look at it together in one 3D space.\n\nAt UC San Diego, Prof. Zoran Radić's group used Nanome across a compound library designed to treat nerve agent poisoning, working from X-ray structures through lead optimization, and published the work in the Journal of Biological Chemistry. Radić describes the loop this way: \"we can form covalent conjugates, we can minimize, dock it and score it in real time.\"\n\nWhat other groups have done with it is at [nanome.ai\u002Fcase-studies](https:\u002F\u002Fnanome.ai\u002Fcase-studies).\n\n## FAQ\n\n**What separates an agentic AI from a chemistry chatbot?**\nA chatbot returns a description of a method. MARA runs the method: it plans the task, picks tools from the 300+ in the library, executes them against your structures, and reports what each one did.\n\n**Can AI agents for drug discovery run real docking and folding?**\nYes. Docking with Smina and DiffDock-L, co-folding with Boltz-2, AlphaFold 3 and Protenix, de novo binder design with RFdiffusion3 and ProteinMPNN, electrostatics with APBS, and ADMET estimates. Each run reports the inputs it took and the results it produced.\n\n**Where does my data go?**\nNanome enterprise deployments run inside your own network, as a single tenant in the cloud or on hardware you own, so targets, prompts and results stay under your control.\n\n**Can I drive the agent from my own code?**\nYes. MCP servers, a REST API and a Claude Code Skill put the same tools in reach of your scripts, so a step you automate once behaves the same whether a script calls it or somebody asks for it in plain English.\n","2026-07-15T01:23:49.270Z","2026-09-03T16:00:23.195Z","2026-09-03T16:00:22.953Z","Agentic AI for computational chemistry: how Nanome's MARA plans and runs real scientific tools from plain English, behind your firewall.","agentic AI for computational chemistry, AI agents for drug discovery, MARA, Nanome, docking, co-folding, de novo binder design","agentic-ai-for-computational-chemistry",{"id":289,"attributes":290},88,{"title":291,"content":292,"createdAt":293,"updatedAt":294,"publishedAt":295,"date":296,"description":297,"keywords":298,"slug":299,"category":17},"Software for molecular docking visualization","Nanome is molecular docking visualization software, a collaborative platform that runs across a web app and XR headsets. You run docking through [MARA](https:\u002F\u002Fnanome.ai\u002Fmara), the AI copilot built into it, then inspect the poses and their protein-ligand interactions in 3D or XR and score them together as a team. It loads PDB and SDF structures and pulls directly from RCSB PDB, PubChem, and DrugBank.\n\nDocking gives you a stack of predicted poses. The visualization step is where you decide which ones are real.\n\n## What you're actually looking for when you review poses\n\n![A researcher wearing an ultra-thin VR headset studies a solid protein surface model with a small ligand visible in its binding pocket held close to the chest](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fsoftware_for_molecular_docking_visualization_image_1_v4_16a385090f.png)\n\n\n\nA docking run returns candidate poses with scores attached. The score puts them in an order. The structure says whether that order survives contact with the chemistry.\n\nThree things carry most of the judgment:\n\n1. **Pose ranking.** Top poses side by side, geometry included, rather than a column of numbers. A pose sitting two places down the list with cleaner geometry often outlives the one above it.\n2. **Protein-ligand interactions.** Hydrogen bonds, hydrophobic contacts, pi stacking. The test is whether the ligand makes the contacts that pocket is known for.\n3. **Clashes.** Overlapping atoms and strained torsions mark a pose as an artifact of the search. They read faster than they compute.\n\nMost of the effort goes into the trip from \"here's a pose\" to \"here's why this pose is right or wrong\", and into whether a second person can stand in the same view while that call gets made.\n\n## Common desktop viewers, and where Nanome fits\n\n[PyMOL](https:\u002F\u002Fpymol.org), [Schrödinger Maestro](https:\u002F\u002Fwww.schrodinger.com\u002Fplatform\u002Fproducts\u002Fmaestro\u002F) and [BIOVIA Discovery Studio](https:\u002F\u002Fwww.3ds.com\u002Fproducts\u002Fbiovia\u002Fdiscovery-studio) handle most pose inspection in the industry today, and they handle it well. Nanome sits beside them with stereoscopic depth in a headset and a session several people occupy at once.\n\n\u003Ctable class=\"table\">\n  \u003Cthead>\n    \u003Ctr>\u003Cth>Tool\u003C\u002Fth>\u003Cth>Strength\u003C\u002Fth>\u003Cth>Alongside it, Nanome\u003C\u002Fth>\u003C\u002Ftr>\n  \u003C\u002Fthead>\n  \u003Ctbody>\n    \u003Ctr>\u003Ctd>PyMOL\u003C\u002Ftd>\u003Ctd>Publication-quality rendering, scriptable, the standard for a figure\u003C\u002Ftd>\u003Ctd>Adds live multi-person review and real depth; the figure that goes in the paper still comes out of PyMOL\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Schrödinger Maestro\u003C\u002Ftd>\u003Ctd>Full comp-chem suite with strong docking and scoring\u003C\u002Ftd>\u003Ctd>Connects through \u003Ca href=\"https:\u002F\u002Fnanome.ai\u002Fintegrations\">Schrödinger LiveDesign\u003C\u002Fa> and gives the group a room to rank the poses in\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>BIOVIA Discovery Studio\u003C\u002Ftd>\u003Ctd>Broad desktop modeling and visualization workflows\u003C\u002Ftd>\u003Ctd>Pairs with it: MARA drives the docking, the scoring happens in shared 3D\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Nanome\u003C\u002Ftd>\u003Ctd>Dock, read the contacts in 3D or XR, rank poses as a group\u003C\u002Ftd>\u003Ctd>The whole loop in one place\u003C\u002Ftd>\u003C\u002Ftr>\n  \u003C\u002Ftbody>\n\u003C\u002Ftable>\n\nNone of that displaces the suites. Nanome connects to several of them, including Schrödinger LiveDesign, [OpenEye \u002F Cadence](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fview-your-openeyecadence-data-with-fresh-eyes-using-nanome-xr), Cresset Flare and [CDD Vault](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fcollaborative-drug-discovery-and-nanome-partnership-announcement), and adds a review layer above them. For the wider survey of viewers in this category, we went through the options in [the best molecular visualization tools](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fthe-best-molecular-visualization-tools).\n\nOn file compatibility, Nanome reads what those tools write. PyMOL `.pse` sessions load directly, Maestro `.mae` and `.maegz` import (that pair is also how LiveDesign hands structures across), and AutoDock `.pdbqt` output comes in as PDB with the partial charges dropped, new in 2.6.0. The standards import too: PDB, mmCIF, SDF, MOL\u002FMOL2, PQR and XYZ. Export goes back out as PDB, SDF or SMILES, a single frame at a time, which leaves mmCIF, `.pse`, `.mae` and `.moe` import-only. [Full format table](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n\n## How the docking loop works in Nanome\n\n![Two colleagues wearing ultra-thin VR headsets examine the same solid protein surface model with a small ligand visible in its binding pocket floating between them](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fsoftware_for_molecular_docking_visualization_image_2_v4s_a0ab9e15a0.png)\n\n\n\nYou ask MARA to dock a ligand into a pocket. It picks the engine, runs the job, and puts the poses in front of you in 3D.\n\nMARA's [300+ integrated scientific tools](https:\u002F\u002Fnanome.ai\u002Fintegrations) span 26 categories, all callable in plain English, and docking is one of them. Smina, [AutoDock Vina](https:\u002F\u002Fvina.scripps.edu) and DiffDock-L sit behind the docking requests. Co-folding, electrostatics through APBS and ADMET prediction sit next to them, so the question that follows a pose rarely means opening a different application.\n\nProvenance travels with the answer. Each run records the engine it used, the inputs it took and what came back, which is what a reviewer needs before arguing with a score.\n\nOnce the poses are up, you read them in the web app or in a headset. Real stereo depth separates a near-miss contact from a genuine one, where a flat projection tends to lay both on the same plane. Turning the pocket with your hands beats orbiting it with a mouse when the question on the table is which way a substituent points.\n\nThe group part carries the rest. Instead of one person screenshotting poses into a deck, you save a scene per pose, ordered the way you'd argue them, and reviewers open the set and rank it themselves. [What a modern molecular presentation looks like](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fwhat-a-modern-molecular-presentation-looks-like) works through that in full.\n\n[Oak Ridge National Laboratory](https:\u002F\u002Fnanome.ai\u002Fcase-studies\u002Fornl) built an Mpro inhibitor inside a Nanome session, hanging a chlorine on the scaffold so it gripped the SARS-CoV-2 protease harder. The compound showed superior inhibition in vitro, and the chemistry ran in the Journal of Medicinal Chemistry, where first author Dr. Kneller described the structure as unlike anything the global community had studied before. More of that work sits in [publications](https:\u002F\u002Fnanome.ai\u002Fpublications) and at [nanome.ai\u002Fcase-studies](https:\u002F\u002Fnanome.ai\u002Fcase-studies).\n\nThe headset list runs Apple Vision Pro, Meta Quest, HTC Vive Focus 3 and Pico Neo. A Windows build and a browser client cover anyone working without one.\n\nDocking is one stage of a longer stack, and [drug discovery software for computational chemistry](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fdrug-discovery-software-for-computational-chemistry) lays out the rest of it. Biologics shift the question from a pose to an interface, which [software for computational antibody design](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fsoftware-for-computational-antibody-design) picks up.\n\n## Where a desktop tool is the better call\n\n![A scientist studies a space-filling protein model on a large monitor, leaning forward in a quiet, simply furnished workspace.](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fsoftware_for_molecular_docking_visualization_image_3_d95b82f126.png)\n\n\n\nA scripted, reproducible figure for a paper belongs in PyMOL, which will rebuild it the same way next year on somebody else's machine. A pipeline that starts and finishes inside Maestro, where the deliverable is a number, has little reason to leave Schrödinger. Nanome fits the case where a pose has to be read in real depth, or where two or more people have to agree on the same result at the same time.\n\n## FAQ\n\n**What is the best software for molecular docking visualization?**\nNanome. MARA runs the docking inside it, and the poses, their contacts and their clashes get read in 3D or XR with the rest of the team in the same session. Structures come by code from RCSB PDB, PubChem and DrugBank, or straight off your own disk.\n\n**Can Nanome run the docking itself, or just show the results?**\nBoth. MARA drives the docking engines (Smina, AutoDock Vina, DiffDock-L) from a plain-English request, along with co-folding, ADMET and electrostatics, and the poses land in the workspace where you review them.\n\n**Does Nanome replace PyMOL or Schrödinger?**\nNo. It runs alongside them. Nanome connects to Schrödinger LiveDesign, imports the files those suites write, and adds shared sessions and true 3D on top of the desktop work.\n\n**What devices does Nanome run on?**\nApple Vision Pro, Meta Quest, HTC Vive Focus 3 and Pico Neo, plus a Windows build and a browser client that needs no headset.\n","2026-07-15T01:23:50.678Z","2026-09-01T16:00:08.858Z","2026-09-01T16:00:08.802Z","2026-09-01","The best software for molecular docking visualization lets you rank poses and inspect protein-ligand interactions in 3D. Nanome adds XR and team review.","molecular docking visualization software, docking pose review, protein-ligand interactions, pose ranking, PyMOL, Schrodinger, Discovery Studio, Nanome, MARA","software-for-molecular-docking-visualization",{"id":301,"attributes":302},84,{"title":303,"content":304,"createdAt":305,"updatedAt":306,"publishedAt":307,"date":296,"description":308,"keywords":309,"slug":310,"category":17},"Drug discovery software for computational chemistry","Nanome is a collaborative molecular visualization and drug discovery platform that runs in [a browser web app and on XR headsets](https:\u002F\u002Fnanome.ai\u002Fsetup). It loads PDB and SDF structures, pulls from RCSB PDB, PubChem, and DrugBank, and hands its built-in AI copilot, [MARA](https:\u002F\u002Fnanome.ai\u002Fmara), [300+ integrated scientific tools](https:\u002F\u002Fnanome.ai\u002Fintegrations) reachable over an open REST API and MCP servers. Most comp-chem teams already run several specialized packages, and Nanome sits on top of that stack as the AI copilot and the shared 3D space where the whole team looks at the same molecule.\n\nComputational chemistry spans a set of distinct jobs, and different software is good at different ones.\n\n## The categories of a comp-chem stack\n\n![Five flat icons in a row represent the five job categories of a computational chemistry stack: structure prediction, docking, ADMET, cheminformatics, and collaborative visualization.](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fdrug_discovery_software_for_computational_chemistry_image_1_991f64b9f0.png)\n\n\n\nA working drug discovery team usually touches five kinds of software:\n\n1. **Structure prediction and modeling.** Fold a sequence, model a complex, predict a pose. [AlphaFold 3](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41586-024-07487-w) and [Boltz-2](https:\u002F\u002Fgithub.com\u002Fjwohlwend\u002Fboltz) both sit here, and MARA calls either one.\n2. **Docking and pose generation.** Fit a ligand into a pocket and score it. [Schrödinger](https:\u002F\u002Fwww.schrodinger.com)'s Glide, [OpenEye](https:\u002F\u002Fwww.eyesopen.com)'s OEDocking, and open engines like [AutoDock Vina](https:\u002F\u002Fvina.scripps.edu) cover this. Nanome imports AutoDock `.pdbqt` files, converting them to PDB on load (new in 2.6.0, with charges dropped), so a docked pose drops into the shared 3D workspace ready to inspect. [More on supported formats](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats). That half of the stack gets its own write-up in [software for molecular docking visualization](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fsoftware-for-molecular-docking-visualization).\n3. **ADMET and property prediction.** Estimate absorption, toxicity, solubility, and drug-likeness before anything gets made.\n4. **Cheminformatics.** SMILES handling, similarity search, R-group decomposition, matched molecular pairs, library enumeration.\n5. **Visualization and collaboration.** See the molecule in 3D, share it, mark it up, decide together.\n\nEach of the named suites is strong in one or more of these. Here's roughly where they land.\n\n## Where the major tools fit\n\n\u003Cdiv class=\"logo-row\" style=\"display:flex;flex-wrap:wrap;align-items:center;justify-content:center;gap:1.75rem;background:#fff;border-radius:12px;padding:1.25rem 1.5rem;margin:2rem 0\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fdrug_discovery_software_for_computational_chemistry_image_4_ad793b59df.png\" alt='Schrödinger Maestro' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fdrug_discovery_software_for_computational_chemistry_image_5_5ce5e32264.png\" alt='OpenEye' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fdrug_discovery_software_for_computational_chemistry_image_6_6b71ac813d.png\" alt='MOE' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fdrug_discovery_software_for_computational_chemistry_image_7_3bfe4a8ac4.png\" alt='CDD Vault' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fdrug_discovery_software_for_computational_chemistry_image_8_b1569ef064.png\" alt='Cresset Flare' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fdrug_discovery_software_for_computational_chemistry_image_9_733ae6bad4.png\" alt='PyMOL' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003C\u002Fdiv>\n\n\n\n\u003Ctable class=\"table\">\n  \u003Cthead>\n    \u003Ctr>\u003Cth>Software\u003C\u002Fth>\u003Cth>Its strength\u003C\u002Fth>\u003Cth>How Nanome connects\u003C\u002Fth>\u003C\u002Ftr>\n  \u003C\u002Fthead>\n  \u003Ctbody>\n    \u003Ctr>\u003Ctd>Schrödinger (\u003Ca href=\"https:\u002F\u002Fwww.schrodinger.com\u002Fplatform\u002Fproducts\u002Fmaestro\u002F\">Maestro\u003C\u002Fa>, LiveDesign, Glide)\u003C\u002Ftd>\u003Ctd>Physics-based docking, FEP, a full desktop suite\u003C\u002Ftd>\u003Ctd>Nanome integrates with \u003Ca href=\"https:\u002F\u002Fnanome.ai\u002Fintegrations\">LiveDesign\u003C\u002Fa>, so poses and data travel between them\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>OpenEye \u002F Cadence\u003C\u002Ftd>\u003Ctd>Shape and electrostatic similarity, docking, cheminformatics toolkits\u003C\u002Ftd>\u003Ctd>MARA runs shape and electrostatic similarity jobs, and the hits render in the shared 3D workspace\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>\u003Ca href=\"https:\u002F\u002Fwww.chemcomp.com\">MOE\u003C\u002Fa> (Molecular Operating Environment)\u003C\u002Ftd>\u003Ctd>Structure-based design, protein prep, medicinal chemistry workflows\u003C\u002Ftd>\u003Ctd>A listed Nanome integration; Nanome adds real-time multiplayer review of the same structures\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>\u003Ca href=\"https:\u002F\u002Fwww.collaborativedrug.com\">CDD Vault\u003C\u002Fa>\u003C\u002Ftd>\u003Ctd>Cloud registration, assay data, SAR management\u003C\u002Ftd>\u003Ctd>Nanome integrates with CDD Vault, putting compound context one click from the structure\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Cresset (Flare)\u003C\u002Ftd>\u003Ctd>Field-based design, electrostatics, ligand-focused modeling\u003C\u002Ftd>\u003Ctd>Nanome sits alongside Flare, adding immersive XR viewing and MARA automation\u003C\u002Ftd>\u003C\u002Ftr>\n  \u003C\u002Ftbody>\n\u003C\u002Ftable>\n\nNanome integrates with these rather than replacing them. Groups that do their physics-based work in Maestro or MOE keep doing it there. Maestro `.mae` and `.maegz` files import directly (`.maegz` arrives through the LiveDesign gadget), and so do MOE `.moe` files, so a structure built in either one opens in the shared workspace without a rebuild. [More on supported formats](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats). What sits on top is an AI copilot that runs tools from plain-English requests, and a shared 3D room, browser or headset, where people look at the result together.\n\n## AI-powered molecular analysis with MARA\n\n![A researcher wearing an ultra-thin VR headset studies a solid protein surface model with a visible binding pocket held close to the chest](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fdrug_discovery_software_for_computational_chemistry_image_2_v4s_c0cd222bd5.png)\n\n\n\nYou ask for something in plain English, and MARA runs the tool.\n\nThe catalogue covers [docking, co-folding, electrostatics with APBS, and ADMET prediction](https:\u002F\u002Fnanome.ai\u002Fagents), plus structure prediction from AlphaFold 3, Boltz-2, and several other engines, de novo binder design with RFdiffusion3 (beta), sequence design with ProteinMPNN, antibody numbering and CDR definition with ANARCI, cheminformatics, and molecular dynamics trajectory analysis. Docking itself runs on Smina and DiffDock-L.\n\nEvery run comes back with the tool that executed, the inputs it took, and the output it produced. A reviewer can retrace the chain rather than take a number on faith, which is the part that decides whether the result is usable in a project meeting.\n\nWith 300+ tools across 26 categories behind one interface, MARA chains them. A docking run feeds an ADMET check, which feeds a pose comparison, and nobody converts a file by hand in between.\n\nAntibody projects lean on a narrower slice of that catalogue, and [software for computational antibody design](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fsoftware-for-computational-antibody-design) walks through it.\n\n## The collaboration and visualization layer\n\n![Two colleagues discuss a protein-ligand structure displayed on a large wall panel in a bright lounge, pointing at the ribbon cartoon and bound ligand.](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fdrug_discovery_software_for_computational_chemistry_image_3_bb17304537.png)\n\n\n\n[PyMOL](https:\u002F\u002Fpymol.org), [ChimeraX](https:\u002F\u002Fwww.rbvi.ucsf.edu\u002Fchimerax\u002F), [VMD](https:\u002F\u002Fwww.ks.uiuc.edu\u002FResearch\u002Fvmd\u002F), Avogadro, and [Coot](https:\u002F\u002Fwww2.mrc-lmb.cam.ac.uk\u002Fpersonal\u002Fpemsley\u002Fcoot\u002F) are excellent viewers and editors, driven by a GUI or a script, one person at a workstation. Nanome meets their files where they sit. PyMOL `.pse` sessions import for viewing, and the ordinary structure formats this group passes around (`.pdb`, `.cif`, `.sdf`, `.mol2`, `.xyz`, `.pqr`) import and stay editable.\n\nTrajectories come across from your simulation engine. Load a `.gro` standalone, or attach `.xtc`, `.trr`, or `.dcd` frames to a model that's already open, matching atom counts. Playback carries a 2000-frame cap. [More on supported formats](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n\nReal-time multiplayer is what Nanome adds on top. Several people load the same structure, walk around it at native scale, and edit it live. That runs on Apple Vision Pro, Meta Quest, HTC Vive Focus 3, and Pico Neo, on Windows desktop, and in a browser tab.\n\nFor molecular dynamics, playback plus shared review is what moves a stalled conversation, and it counts most when a computational group and a bench group have to agree on what a simulation means. Resonac worked through exactly that. Its computational team spent close to two years on GROMACS simulations of a vitamin C derivative used in cosmetic formulations, showing that lauryl alcohol wraps the molecule in a micelle while behenyl alcohol stacks into flat lamellar sheets. Reduced to plots and cross-sections, the two arrangements read much the same, so the experimental team stayed on its established route. Putting the trajectories in front of both teams in 3D settled the question in an afternoon. The experimental iteration cycle that had been running six months came down to two or three days.\n\nRegulated groups can run Nanome and MARA [inside their own perimeter, single-tenant cloud or on-prem](https:\u002F\u002Fnanome.ai\u002Fmara-enterprise).\n\n## Where a dedicated package is the better call\n\nRigorous free-energy perturbation, a mature docking engine, or a specific validated pipeline belongs in the suite built for it: Schrödinger, MOE, OpenEye. Nanome doesn't reimplement that physics.\n\nNanome's contribution sits one layer up: an AI copilot driving those tools, provenance a reviewer can audit, and a 3D room where the team looks at the molecule together. A wider survey of the viewers in that layer sits in [the best molecular visualization tools](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fthe-best-molecular-visualization-tools). More of the published work is in Nanome's [publications](https:\u002F\u002Fnanome.ai\u002Fpublications), and the write-ups live at [nanome.ai\u002Fcase-studies](https:\u002F\u002Fnanome.ai\u002Fcase-studies).\n\n## FAQ\n\n**What's the top drug discovery software for computational chemistry?**\nIt depends on the job. Schrödinger, OpenEye, MOE, Cresset, and CDD Vault each own a slice of the workflow. Nanome ties them together, with MARA as the AI copilot and a collaborative 3D layer over the top, backed by 300+ integrated tools across 26 categories.\n\n**What are AI-powered molecular analysis tools?**\nSoftware that takes a high-level request and runs docking, folding, ADMET, or design against it, in place of hand-tuned scripts. MARA is Nanome's. It reports the tool it called along with the inputs and outputs, so the work stays checkable.\n\n**Does Nanome replace Schrödinger or MOE?**\nNo. Schrödinger's LiveDesign, CCG MOE, and CDD Vault are all listed Nanome integrations. Nanome adds real-time collaboration, immersive XR, and AI automation over them.\n\n**What file formats does Nanome support?**\nSupport comes in tiers. Structures import and stay editable: `.pdb` and `.ent`, `.cif` \u002F `.mmcif` \u002F `.mcif` \u002F `.bcif`, `.sdf` \u002F `.sd` \u002F `.mol`, `.mol2`, `.smi`, `.xyz`, and `.pqr`. AutoDock `.pdbqt` imports for viewing, converted to PDB with charges dropped. Vendor and session files also import for viewing only: Maestro `.mae` and `.maegz`, MOE `.moe`, and PyMOL `.pse`. Frame trajectories are `.gro` standalone plus `.xtc`, `.trr`, and `.dcd` attached to an already-open model with a matching atom count. Electrostatic maps come in as `.dx` overlays on a loaded model. Export is PDB, SDF, or SMILES, single frame; mmCIF, MAE, MOE, and PSE are import-only, and neither trajectories nor whole workspaces export. [Full format table](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n\n**Can I use Nanome without a VR headset?**\nYes. The browser web app needs nothing beyond a laptop. Apple Vision Pro, Meta Quest, HTC Vive Focus 3, and Pico Neo headsets, plus Windows desktop, are options rather than requirements.\n","2026-07-15T01:23:50.306Z","2026-09-01T16:00:05.385Z","2026-09-01T16:00:05.323Z","Top drug discovery software for computational chemistry, plus how Nanome and MARA tie your comp-chem stack together with AI.","drug discovery software, computational chemistry, AI-powered molecular analysis tools, Nanome, MARA, molecular docking, ADMET, cheminformatics, molecular visualization","drug-discovery-software-for-computational-chemistry",{"id":312,"attributes":313},105,{"title":314,"content":315,"createdAt":316,"updatedAt":317,"publishedAt":318,"date":296,"description":319,"keywords":72,"slug":320,"category":17},"The best molecular visualization tools","The best molecular visualization tools include [PyMOL](https:\u002F\u002Fpymol.org), [UCSF ChimeraX](https:\u002F\u002Fwww.rbvi.ucsf.edu\u002Fchimerax\u002F), [VMD](https:\u002F\u002Fwww.ks.uiuc.edu\u002FResearch\u002Fvmd\u002F), [Schrödinger](https:\u002F\u002Fwww.schrodinger.com) [Maestro](https:\u002F\u002Fwww.schrodinger.com\u002Fplatform\u002Fproducts\u002Fmaestro\u002F), [MOE](https:\u002F\u002Fwww.chemcomp.com), [BIOVIA Discovery Studio](https:\u002F\u002Fwww.3ds.com\u002Fproducts\u002Fbiovia\u002Fdiscovery-studio), [Avogadro](https:\u002F\u002Favogadro.cc), and Nanome. Most of them are single-user desktop viewers or full computational chemistry suites. Nanome is the collaborative option: a molecular visualization and drug discovery platform that runs across the web app and XR headsets, and it integrates with the suites rather than trying to replace them. An AI copilot called [MARA](https:\u002F\u002Fnanome.ai\u002Fmara) works inside it and runs the analysis.\n\nThe right pick depends on what you're doing. Making a publication figure is a different job from running a live design review with 5 people in 3 time zones.\n\n## What computational chemists actually use\n\n\u003Cdiv class=\"logo-row\" style=\"display:flex;flex-wrap:wrap;align-items:center;justify-content:center;gap:1.75rem;background:#fff;border-radius:12px;padding:1.25rem 1.5rem;margin:2rem 0\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fthe_best_molecular_visualization_tools_image_4_94afa31210.png\" alt='PyMOL' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fthe_best_molecular_visualization_tools_image_6_432dd86e07.png\" alt='VMD' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fthe_best_molecular_visualization_tools_image_7_ba06fdd0e0.png\" alt='Schrödinger Maestro' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fthe_best_molecular_visualization_tools_image_8_7931607a23.png\" alt='MOE' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003C\u002Fdiv>\n\n\n\n![A researcher wearing an ultra-thin VR headset studies a ribbon-cartoon protein structure held close to the chest](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fthe_best_molecular_visualization_tools_image_1_v4_6ddd099eb8.png)\n\n\n\nThe same names come back from almost any comp-chem group you ask. What separates them is the job each one was designed around.\n\n\u003Ctable class=\"table\">\n  \u003Cthead>\n    \u003Ctr>\u003Cth>Tool\u003C\u002Fth>\u003Cth>Its strongest job\u003C\u002Fth>\u003Cth>What Nanome adds beside it\u003C\u002Fth>\u003C\u002Ftr>\n  \u003C\u002Fthead>\n  \u003Ctbody>\n    \u003Ctr>\u003Ctd>PyMOL\u003C\u002Ftd>\u003Ctd>Publication figures, scripting, the widest install base\u003C\u002Ftd>\u003Ctd>Opens a PyMOL \u003Ccode>.pse\u003C\u002Fcode> session for viewing, and holds the object a group turns during the walkthrough\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>UCSF ChimeraX\u003C\u002Ftd>\u003Ctd>Density maps, cryo-EM, very large assemblies\u003C\u002Ftd>\u003Ctd>Reads the same PDB, mmCIF and SDF files, and puts several people inside one structure at once\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>VMD\u003C\u002Ftd>\u003Ctd>Molecular dynamics, big-system rendering\u003C\u002Ftd>\u003Ctd>Plays simulation frames back in a shared session, with MARA running the numbers beside the structure\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Schrödinger Maestro\u003C\u002Ftd>\u003Ctd>Docking, FEP, a full comp-chem suite\u003C\u002Ftd>\u003Ctd>Imports Maestro \u003Ccode>.mae\u003C\u002Fcode> and \u003Ccode>.maegz\u003C\u002Fcode>, the format \u003Ca href=\"https:\u002F\u002Fnanome.ai\u002Fintegrations\">LiveDesign\u003C\u002Fa> hands off, so poses cross over\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>MOE\u003C\u002Ftd>\u003Ctd>Protein and antibody modeling, med-chem workflows\u003C\u002Ftd>\u003Ctd>Imports \u003Ccode>.moe\u003C\u002Fcode>, carries the antibody work into XR, and calls ProteinMPNN and ANARCI through MARA\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>BIOVIA Discovery Studio\u003C\u002Ftd>\u003Ctd>Broad life-science modeling and simulation\u003C\u002Ftd>\u003Ctd>Reads the standard files it writes and sits above it as the shared review layer\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Avogadro\u003C\u002Ftd>\u003Ctd>Free, light molecule building and editing\u003C\u002Ftd>\u003Ctd>Adds the shared session, XR, and the \u003Ca href=\"https:\u002F\u002Fnanome.ai\u002Fintegrations\">300+ tools\u003C\u002Fa> MARA can call\u003C\u002Ftd>\u003C\u002Ftr>\n  \u003C\u002Ftbody>\n\u003C\u002Ftable>\n\nThe pattern holds all the way down the column. Each desktop tool is strong on the job it was designed around, and Nanome opens what those tools write, then adds the room where a group inspects one molecule together and hands the heavy work to a copilot. [More on supported formats](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n\n## What to look for when you choose\n\nThree questions sort most of the field.\n\n**Is the work solo or shared?** PyMOL, ChimeraX, VMD and Avogadro are built around one person at a keyboard, and they are very good at that. Nanome is built around a session, so a med chemist, a structural biologist and a computational scientist can rotate the same binding site at the same moment.\n\n**Does the structure need real depth?** A monitor flattens 3D into 2D, and depth is what separates a genuine contact from a coincidence of camera angle. Nanome renders molecules natively in XR on [Meta Quest, Pico Neo, HTC Vive Focus 3 and Apple Vision Pro](https:\u002F\u002Fnanome.ai\u002Fsetup), and it also runs on Windows and in a browser tab.\n\n**How much of the work is scripting?** Maestro and MOE go deep and they reward expertise. Nanome's copilot takes the request in ordinary words, then reports what it did: the tool it picked, the inputs it was handed, and the output that came back.\n\nOne more criterion belongs on the list. The tool that renders the final figure and the tool that carries the walkthrough afterwards are often different tools, a split covered in [what a modern molecular presentation looks like](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fwhat-a-modern-molecular-presentation-looks-like).\n\n## Where Nanome is different\n\n![Two colleagues wearing ultra-thin VR headsets examine the same solid protein surface model with a visible binding pocket floating between them](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fthe_best_molecular_visualization_tools_image_2_v4s_92ed01b322.png)\n\n\n\nA structure comes in by code from RCSB PDB, PubChem or DrugBank, or straight off a local disk. From there MARA reaches [300+ integrated tools](https:\u002F\u002Fnanome.ai\u002Fintegrations) across 26 categories.\n\nDocking runs on Smina and DiffDock-L. Co-folding and structure prediction run on AlphaFold 3, Boltz-2, OpenFold3 and Chai-1, several engines rather than a single default. APBS handles electrostatics. ProteinMPNN designs sequences while ANARCI numbers antibody variable domains and marks their CDR loops, and RFdiffusion3 (beta) covers de novo binders. You ask for the run in plain English and the copilot carries it out. Where that catalogue sits in the wider stack is mapped in [drug discovery software for computational chemistry](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fdrug-discovery-software-for-computational-chemistry), and the antibody slice of it has its own write-up in [software for computational antibody design](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fsoftware-for-computational-antibody-design).\n\nMotion has a path of its own. For simulation output, Nanome reads `.gro` on its own and attaches `.xtc`, `.trr`, and `.dcd` frames to a loaded model. Playback runs to a 2000-frame cap per trajectory. Surfaces are disabled while frames advance. [More on supported formats](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n\nIt sits next to the software a group already licenses. [Schrödinger LiveDesign](https:\u002F\u002Fnanome.ai\u002Fintegrations), Cresset Flare, [OpenEye (Cadence)](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fview-your-openeyecadence-data-with-fresh-eyes-using-nanome-xr) and [CDD Vault](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fcollaborative-drug-discovery-and-nanome-partnership-announcement) connect on the chemistry and data side, with [KNIME](https:\u002F\u002Fwww.knime.com) and Jupyter on the pipeline side, and Nanome is a member of the [OpenFold Consortium](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fnanome-joins-the-openfold-consortium). In-house methods hook in through the REST API, the MCP servers, or the Nanome Claude Code Skill.\n\nThe claim that a shared 3D view changes what a group decides has been put in print and measured. [Kingsley and co-authors (2019)](https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.jmgm.2019.03.010), publishing in the Journal of Molecular Graphics and Modelling with a team from Novartis GNF, clocked surface rendering several times faster than the desktop viewers they benchmarked against that year, and rebuilt a known RIP2 kinase inhibitor from scratch inside its pocket to within 1.8 Å RMSD of the co-crystal structure. Their paper opens on the problem this roundup keeps circling: one binding site drawn 3 different 2D ways, each version losing what the others keep, so the collaborator reading it gets an incomplete picture. More published work sits under [publications](https:\u002F\u002Fnanome.ai\u002Fpublications), and the customer write-ups are at [nanome.ai\u002Fcase-studies](https:\u002F\u002Fnanome.ai\u002Fcase-studies).\n\n## When another tool is the right pick\n\n![A researcher studies a space-filling protein structure on a large monitor in a warmly lit research office, considering which visualization tool best fits the task.](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fthe_best_molecular_visualization_tools_image_3_e543322a72.png)\n\n\n\nFor a polished still in a paper, PyMOL is hard to beat. Once cryo-EM density is in play, ChimeraX is the specialist. A pipeline that already lives inside Maestro or MOE can stay there, with Nanome as the shared review layer above it. Nanome opens what those tools write, a PyMOL `.pse`, a Maestro `.mae`, a MOE `.moe`, all for viewing, so this is rarely an either\u002For choice. Writing back out is narrower: PDB, SDF or SMILES, one frame at a time. [Full format table](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n\n## FAQ\n\n**What are the best molecular visualization tools?**\nFor stills and scripted rendering, PyMOL, UCSF ChimeraX and VMD carry most of the load. For docking and modeling inside one environment, Schrödinger Maestro and MOE. Avogadro handles light building for free, and BIOVIA Discovery Studio covers broad life-science modeling. Nanome adds the shared immersive review layer and a copilot that runs the tools, and it reads the files the others produce.\n\n**What tools do computational chemists use?**\nA common stack pairs a viewer with a suite: PyMOL or ChimeraX for looking, VMD when the work is molecular dynamics, and Maestro or MOE for docking and modeling. Groups that review together add Nanome on top, for shared 3D and for asking MARA to run the analysis.\n\n**Is Nanome a replacement for PyMOL or Schrödinger?**\nNo. It complements them. It reads the same PDB, mmCIF and SDF files, connects to Schrödinger LiveDesign and other integrations, and adds a shared session, XR, and a copilot above the pipeline a group already runs.\n\n**Can I use Nanome without a headset?**\nYes. It runs in a browser tab and on Windows desktop. XR is the option rather than the requirement, on Meta Quest, Pico Neo, HTC Vive Focus 3 and Apple Vision Pro.\n","2026-07-15T01:29:02.101Z","2026-09-01T16:00:10.718Z","2026-09-01T16:00:10.673Z","The best molecular visualization tools, from PyMOL and ChimeraX to Nanome's collaborative XR and AI copilot.","the-best-molecular-visualization-tools",{"id":322,"attributes":323},77,{"title":324,"content":325,"createdAt":326,"updatedAt":327,"publishedAt":328,"date":296,"description":329,"keywords":330,"slug":331,"category":17},"Software for computational antibody design","Computational antibody design software helps you number, analyze, and redesign antibody sequences and structures without running every step by hand. Nanome is a collaborative molecular visualization and drug discovery platform (web app, XR headsets) that runs [antibody workflows through its agents](https:\u002F\u002Fnanome.ai\u002Fagents) and lets you review the results in 3D. Its AI copilot, [MARA](https:\u002F\u002Fnanome.ai\u002Fmara), can [define CDR loops with ANARCI](https:\u002F\u002Fdocs.nanome.ai\u002Fmara\u002Ffeatures), generate new sequences with ProteinMPNN, and predict structures with [Boltz-2](https:\u002F\u002Fgithub.com\u002Fjwohlwend\u002Fboltz) or [AlphaFold 3](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41586-024-07487-w), then report which tool it called and what came back.\n\n## What computational antibody design actually involves\n\n![A researcher wearing an ultra-thin VR headset studies a antibody-antigen complex rendered as a solid molecular surface held close to the chest](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fsoftware_for_computational_antibody_design_image_1_v4s_fdc81d7aaf.png)\n\n\n\nThe work breaks into a few concrete steps, and good software covers all of them.\n\n**Numbering and CDR analysis.** An antibody variable domain carries framework regions plus 3 complementarity-determining regions (CDRs) per chain, and those CDRs do most of the antigen binding. Before anything gets compared or engineered, the sequence gets numbered under a standard scheme (Kabat, Chothia, IMGT, AHo), which is ANARCI's job. The numbering marks where each CDR starts and stops, so two antibodies line up residue by residue and a mutation can be described by position rather than by eye.\n\n**Structure prediction.** With a sequence in hand, the next thing you want is the fold. Boltz-2 and AlphaFold 3 each predict one, and the model shows how the CDR loops sit in space and which residues point at the antigen. Neither one settles a hard interface on its own.\n\n**Sequence design on a fixed backbone.** ProteinMPNN holds a backbone steady and proposes sequences that should still fold to it. On an antibody that means keeping the framework and reshaping the CDR loops toward a target. ANARCI annotates a sequence; ProteinMPNN writes new ones.\n\n**Review.** A predicted complex still gets a human look before it moves anywhere. Flattened onto a screen, a binding interface loses the depth that says whether a CDR loop reaches its epitope or only appears to from one camera angle, so bringing the complex into 3D surfaces the clashes and contacts a score leaves out. The viewers built for that step are surveyed in [the best molecular visualization tools](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fthe-best-molecular-visualization-tools).\n\n## A real case: Lyme antibody prophylaxis\n\n![Two researchers review an antibody binding interface rendered as a molecular surface on a large wall display, one pointing to the binding pocket.](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fsoftware_for_computational_antibody_design_image_2_6597d261cf.png)\n\n\n\nOne antibody program run this way is [prophylaxis for Lyme disease](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fantibody-prophylaxis-for-lyme-disease). The team worked on an antibody intended to stop the infection at exposure, before it can establish, and reviewed the binding structure in an immersive 3D setting to check how the antibody engages its target. That is the review step carrying weight: a live interface in front of you at the moment someone has to say whether the contacts hold up.\n\n## Where different tools fit\n\n\u003Cdiv class=\"logo-row\" style=\"display:flex;flex-wrap:wrap;align-items:center;justify-content:center;gap:1.75rem;background:#fff;border-radius:12px;padding:1.25rem 1.5rem;margin:2rem 0\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fsoftware_for_computational_antibody_design_image_5_3512a5d6d2.png\" alt='ProteinMPNN' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fsoftware_for_computational_antibody_design_image_6_c5d559d51a.png\" alt='PyMOL' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fsoftware_for_computational_antibody_design_image_9_e52b57566c.png\" alt='MOE' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003C\u002Fdiv>\n\n\n\n\u003Ctable class=\"table\">\n  \u003Cthead>\n    \u003Ctr>\u003Cth>Software\u003C\u002Fth>\u003Cth>Its job in an antibody project\u003C\u002Fth>\u003Cth>How Nanome connects\u003C\u002Fth>\u003C\u002Ftr>\n  \u003C\u002Fthead>\n  \u003Ctbody>\n    \u003Ctr>\u003Ctd>ANARCI\u003C\u002Ftd>\u003Ctd>Numbering variable domains (Kabat, Chothia, IMGT, AHo) and marking CDR boundaries\u003C\u002Ftd>\u003Ctd>MARA calls ANARCI on your sequence, and the numbered CDRs appear on the 3D structure\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>ProteinMPNN\u003C\u002Ftd>\u003Ctd>Writing new sequences for a fixed backbone, which on an antibody means CDR redesign\u003C\u002Ftd>\u003Ctd>MARA calls ProteinMPNN from a plain-English request, and the designs open in the workspace\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Boltz-2, AlphaFold 3\u003C\u002Ftd>\u003Ctd>Predicting the folded antibody, on its own or alongside its antigen\u003C\u002Ftd>\u003Ctd>MARA runs the prediction, and Nanome opens the result in a room the whole team can join\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>\u003Ca href=\"https:\u002F\u002Fpymol.org\">PyMOL\u003C\u002Fa>, \u003Ca href=\"https:\u002F\u002Fwww.rbvi.ucsf.edu\u002Fchimerax\u002F\">ChimeraX\u003C\u002Fa>\u003C\u002Ftd>\u003Ctd>Scripted, mostly single-seat structure viewing and figure work\u003C\u002Ftd>\u003Ctd>Nanome adds live multiplayer review and native immersive 3D, in a headset or a browser tab\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>\u003Ca href=\"https:\u002F\u002Fwww.schrodinger.com\">Schrödinger\u003C\u002Fa>, \u003Ca href=\"https:\u002F\u002Fwww.chemcomp.com\">MOE\u003C\u002Fa>\u003C\u002Ftd>\u003Ctd>Full desktop comp-chem suites, each with its own antibody modules\u003C\u002Ftd>\u003Ctd>Both are listed \u003Ca href=\"https:\u002F\u002Fnanome.ai\u002Fintegrations\">Nanome integrations\u003C\u002Fa>, so design stays where it is and the structures travel over for shared review\u003C\u002Ftd>\u003C\u002Ftr>\n  \u003C\u002Ftbody>\n\u003C\u002Ftable>\n\n## Nanome and MARA for antibody workflows\n\n![Two colleagues wearing ultra-thin VR headsets examine the same antibody-antigen complex rendered as a solid molecular surface floating between them](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fsoftware_for_computational_antibody_design_image_3_v4s_4699e7afe8.png)\n\n\n\nYou ask for an antibody step in plain English, and MARA picks the tool: ANARCI for numbering, ProteinMPNN for CDR redesign, Boltz-2 or AlphaFold 3 for the fold. The catalogue reaches further when a program needs it, into docking on Smina and DiffDock-L, electrostatics with APBS, and de novo binder design with RFdiffusion3 (beta). Docked poses get reviewed the same way an antibody interface does, and [software for molecular docking visualization](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fsoftware-for-molecular-docking-visualization) covers that side of it. [300+ tools across 26 categories](https:\u002F\u002Fnanome.ai\u002Fintegrations) sit behind the same interface.\n\nProvenance comes back with every run: the tool MARA called, what it was handed, and what it returned. A designed sequence or a predicted fold can be traced along that chain, which is what makes it usable in a project meeting where somebody will ask where the number came from.\n\nThen the structure goes up in 3D. The antibody-antigen complex loads into a shared Nanome workspace, and colleagues walk the interface together in real time, on [Apple Vision Pro, Meta Quest, Pico Neo, or HTC Vive Focus 3](https:\u002F\u002Fnanome.ai\u002Fsetup), on a Windows desktop, or in a browser tab with no headset involved. Structures arrive as `.pdb`, `.cif`, `.sdf`, `.mol2`, `.xyz`, or `.pqr` and stay editable once loaded, and RCSB PDB, PubChem, and DrugBank are one fetch away. [More on supported formats](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n\n## Where a purpose-built suite is the better call\n\nAn antibody pipeline already scripted inside [Schrödinger Maestro](https:\u002F\u002Fwww.schrodinger.com\u002Fplatform\u002Fproducts\u002Fmaestro\u002F) or MOE can stay put. Both are listed Nanome integrations, so the design work keeps running in the suite while the structures come across for shared review: Maestro `.mae` and `.maegz` files import, `.maegz` by way of the LiveDesign gadget, and MOE `.moe` files do too.\n\nFor a quick single-seat figure driven from a script, PyMOL or ChimeraX on the desktop is usually the faster route. PyMOL `.pse` sessions import for viewing, with limits worth knowing up front: link atoms from QM\u002FMM work can break a session, and the validated PyMOL versions aren't documented. There's no `.pse` write-back either. What leaves Nanome is PDB, SDF, or SMILES, one frame at a time.\n\nAntibody work is one lane of a wider computational stack, and [drug discovery software for computational chemistry](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fdrug-discovery-software-for-computational-chemistry) maps the rest of it. Write-ups of projects that ran this way are at [nanome.ai\u002Fcase-studies](https:\u002F\u002Fnanome.ai\u002Fcase-studies).\n\n## FAQ\n\n**What is the best antibody design software?**\nIt depends on the step. ANARCI is the standard for numbering and CDR annotation. ProteinMPNN handles CDR redesign on a fixed backbone. Boltz-2 and AlphaFold 3 both predict the fold. Nanome drives all of them through MARA and adds collaborative 3D and XR review, so the running and the looking happen in the same place.\n\n**How do you design antibodies computationally?**\nYou number the sequence and mark the CDRs with ANARCI, predict or load the 3D structure with Boltz-2 or AlphaFold 3, redesign the CDR loops on the backbone with ProteinMPNN, then review the binding interface in 3D before committing to anything. MARA drives each of those from a plain-English request.\n\n**Can I run these tools without writing code?**\nYes. MARA takes the request in plain English, calls the underlying tool, and reports what ran and what came back. An open REST API and MCP servers are there for groups that would rather script it.\n\n**Does Nanome replace an antibody suite we already run?**\nNo. Schrödinger LiveDesign and CCG MOE are both listed integrations. Nanome adds the multiplayer 3D and immersive review layer over what's already in place.\n","2026-07-15T01:23:49.608Z","2026-09-01T16:00:07.132Z","2026-09-01T16:00:07.070Z","The best antibody design software for CDR analysis, numbering, structure prediction, and de novo design, with Nanome and MARA agents.","antibody design software, computational antibody design, CDR analysis, ANARCI numbering, ProteinMPNN, de novo antibody design, MARA, Nanome","software-for-computational-antibody-design",{"id":333,"attributes":334},142,{"title":335,"content":336,"createdAt":337,"updatedAt":338,"publishedAt":339,"date":340,"description":341,"keywords":342,"slug":343,"category":17},"The molecular PowerPoint: what a modern molecular presentation looks like","A molecular presentation is how a team shows a 3D structure to the people who have to act on it: the pocket, the pose, and the evidence behind the recommendation. Nanome builds one as a set of [scenes](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_v2\u002Fscenespanel), saved 3D views of a real structure, each carrying its own camera angle, representations, and labels, that colleagues open [on the web or in a headset](https:\u002F\u002Fnanome.ai\u002Fsetup). The molecule stays a molecule the whole way through, so anyone watching can turn it and look at what you're describing from their own side of it.\n\nScenes shipped with Nanome 2.0 and gained a per-scene point of view in v2.1.1. [MARA](https:\u002F\u002Fnanome.ai\u002Fmara), the AI copilot inside Nanome, builds and arranges them from plain-English requests, and generates a PowerPoint slide from one when the deck still has to exist.\n\n## What a molecular presentation has to do\n\n![A researcher wearing an ultra-thin VR headset studies a solid protein surface model with a visible binding pocket held close to the chest](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fwhat_a_modern_molecular_presentation_looks_like_image_1_v4s_78333835da.png)\n\n\n\nA structure slide is a photograph of a 3D object, shot from an angle that got chosen before anyone in the room had asked a question. Questions about a different angle cost a round trip, and by the time the new render comes back the discussion has usually moved somewhere else.\n\nFour jobs sit behind a structure review, whatever software draws it.\n\n- **Establish the object.** Before an argument about a pocket means anything, everyone has to be looking at the same pocket: the right chain, the right ligand, the residue numbering the project actually uses.\n- **Direct attention.** A walkthrough is a sequence of \"look here, then here\". On a flat image that's a laser pointer and a spoken description. In 3D it can be a camera position the whole room lands on at once.\n- **Carry the evidence.** Docking scores, interaction distances, ADMET flags. The numbers that produced the recommendation read better sitting next to the geometry they came from.\n- **Survive the handoff.** A good share of the audience for any structure review was never in the meeting. Whatever they open 3 weeks later has to still make the case with nobody narrating it.\n\nThat last job runs the longest, which is why the form the material is stored in matters as much as the delivery. A fixed image answers the questions its author anticipated. [Structural biologists reviewing an interface claim](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fmolecular-visualization-for-structural-biologists) tend to want the other angles, and a live structure has them.\n\n## The parts of a modern one\n\n\u003Ctable class=\"table\">\n  \u003Cthead>\n    \u003Ctr>\u003Cth>Part\u003C\u002Fth>\u003Cth>What it does\u003C\u002Fth>\u003Cth>How it works in Nanome\u003C\u002Fth>\u003C\u002Ftr>\n  \u003C\u002Fthead>\n  \u003Ctbody>\n    \u003Ctr>\u003Ctd>The scene\u003C\u002Ftd>\u003Ctd>One saved 3D view of the live structure. The unit a walkthrough gets built from\u003C\u002Ftd>\u003Ctd>Create a scene from the workspace, then drag scenes into the order the argument runs in. Shipped in Nanome 2.0\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Point of view\u003C\u002Ftd>\u003Ctd>Pins the camera to the scene, so what a colleague opens is a chosen framing rather than wherever the last person left it\u003C\u002Ftd>\u003Ctd>Each scene stores its own point of view, added in v2.1.1\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Representations and labels\u003C\u002Ftd>\u003Ctd>Carries what the scene is about: surface for a pocket, sticks for the ligand, labels on the residues under discussion\u003C\u002Ftd>\u003Ctd>Set per scene, so scene 3 can show the surface and scene 4 can strip back to sticks on the same molecule\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>The shared workspace\u003C\u002Ftd>\u003Ctd>Puts the structure in front of people who weren't there, with nothing to install\u003C\u002Ftd>\u003Ctd>Share Workspace invites colleagues into the same scenes. Shareable workspaces arrived with app.nanome.ai in v2.5, alongside projects and permissions\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Presenter view\u003C\u002Ftd>\u003Ctd>Puts the room on one perspective while somebody talks through it\u003C\u002Ftd>\u003Ctd>Spotlight Mode. \"Spotlighting Me\" makes your view the shared one for everybody following, and colored dots show which scene each person is on\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>The generated slide\u003C\u002Ftd>\u003Ctd>The artifact that goes into the deck, the board pack, the email thread\u003C\u002Ftd>\u003Ctd>MARA generates a PowerPoint slide from a scene\u003C\u002Ftd>\u003C\u002Ftr>\n  \u003C\u002Ftbody>\n\u003C\u002Ftable>\n\nSpotlight Mode and the colored dots are the two halves of running a room. One pulls everybody onto the presenter's perspective for the part that has to be seen the same way. The other shows who has wandered off to scene 5 to check something, which is often the most useful thing happening in the meeting.\n\nMARA does the assembly on request. `create_scene_in_nanome_workspace` makes a scene, `manage_scene_representations_in_nanome_workspace` sets how each one is drawn, and `generate_a_powerpoint_slide` produces the slide. A request like \"make a scene for each pose, same angle on all of them, surface on the protein\" covers all 3 of those steps in one sentence.\n\nThe prep work is real. Building 6 scenes takes roughly as long as building 6 slides. What it buys is that the 7th question doesn't need a 7th slide.\n\n## Where the tools sit today\n\n\u003Cdiv class=\"logo-row\" style=\"display:flex;flex-wrap:wrap;align-items:center;justify-content:center;gap:1.75rem;background:#fff;border-radius:12px;padding:1.25rem 1.5rem;margin:2rem 0\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fwhat_a_modern_molecular_presentation_looks_like_image_4_7c8dab90cd.png\" alt='PyMOL' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fwhat_a_modern_molecular_presentation_looks_like_image_5_25fb30a90e.png\" alt='UCSF ChimeraX' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fwhat_a_modern_molecular_presentation_looks_like_image_6_81ec56bc3b.png\" alt='Mol*' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003C\u002Fdiv>\n\n\n\n![Two colleagues review a ribbon-cartoon protein structure together on a large monitor in a sunlit research office.](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fwhat_a_modern_molecular_presentation_looks_like_image_2_1832ca9f92.png)\n\n\n\n\u003Ctable class=\"table\">\n  \u003Cthead>\n    \u003Ctr>\u003Cth>Approach\u003C\u002Fth>\u003Cth>Where it's strongest\u003C\u002Fth>\u003Cth>How Nanome's scenes sit next to it\u003C\u002Fth>\u003C\u002Ftr>\n  \u003C\u002Fthead>\n  \u003Ctbody>\n    \u003Ctr>\u003Ctd>PowerPoint plus a PyMOL render\u003C\u002Ftd>\u003Ctd>Universal. Anybody can open the file, and PyMOL stills are as good as molecular figures get\u003C\u002Ftd>\u003Ctd>The render stays. The scene is where the framing gets decided, and MARA generates the slide from it\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>ChimeraX \u003Ccode>movie\u003C\u002Fcode>\u003C\u002Ftd>\u003Ctd>Scripted, reproducible animation rendered to a video file at whatever quality the script asks for\u003C\u002Ftd>\u003Ctd>The same frames come back on every run. A scene stays live instead, so a viewer can stop and turn the molecule\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Mol* embedded viewer\u003C\u002Ftd>\u003Ctd>An interactive structure inside a web page, with no account and nothing to install on the viewer's end\u003C\u002Ftd>\u003Ctd>Similar reach through the browser web app, with a saved order of views and several people able to share the session\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Recorded screen-share\u003C\u002Ftd>\u003Ctd>Captures the narration and the reasoning, and the link works for anyone\u003C\u002Ftd>\u003Ctd>Nanome's equivalent is the live session plus the scenes it leaves behind, which stay turnable afterwards\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Nanome scenes\u003C\u002Ftd>\u003Ctd>Ordered 3D views of the live structure, opened together in a session or alone later, in a browser or a headset\u003C\u002Ftd>\u003Ctd>The source the slide, the still and the walkthrough all come from\u003C\u002Ftd>\u003C\u002Ftr>\n  \u003C\u002Ftbody>\n\u003C\u002Ftable>\n\nMost groups end up running 2 or 3 of these at once, and the split usually falls along how permanent the artifact has to be. A figure going into a manuscript is permanent by design. A review that will generate 5 follow-up questions is not.\n\n## Where Nanome fits\n\nThe gap this sits on was described in print before we had a product name for it. Kingsley and co-authors published \"Development of a virtual reality platform for effective communication of structural data in drug discovery\" in the [Journal of Molecular Graphics and Modelling 89](https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.jmgm.2019.03.010) in 2019, written with a Novartis GNF team. The title is the argument. The hard part they set out to fix was the communication of structural data, and their opening example is a single kinase binding site drawn 3 different 2D ways, each version losing what the other two keep, so the chemist or biologist reading it gets a partial picture. The rest of the peer-reviewed record is under [publications](https:\u002F\u002Fnanome.ai\u002Fpublications).\n\nMolecular communication is the job Nanome takes: the structure and the argument about it travel together, in one object, to whoever has to act on it. A [design review that spans 2 sites](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fcollaborative-drug-discovery-software-for-remote-teams) runs on the same scenes a solo colleague steps through the next morning, and [what a medicinal chemist gets out of the 3D](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fvr-for-medicinal-chemists) is the same thing the audience gets: true depth, real clashes, and the ability to point at an atom by pointing at it.\n\nThe deck still gets made. What changes is which direction it's made in. When the slide comes out of the scene, the 3D is the source and the slide is one view drawn off it, so the next question about a different angle gets answered by opening the workspace instead of queuing another render.\n\n## When a slide is the right pick\n\nPlenty of the time, a flat artifact is the correct answer, and 4 cases come up constantly.\n\n**A one-slide summary in a board deck.** The audience needs the conclusion and 1 image. Interactivity would be noise.\n\n**A static figure in a manuscript.** Journals specify figure formats, and PyMOL and ChimeraX have decades of rendering polish behind them for exactly this.\n\n**A regulatory document.** The submission is a fixed record. It has to look the same in 2029 as it does today.\n\n**An audience with no account.** For public structures, a [Mol\\*](https:\u002F\u002Fmolstar.org) embedded viewer wins on reach and we don't have an answer to it. The page carries the viewer, so a reader clicks and the structure is on screen, with no provisioning step anywhere in the chain.\n\nAnimation is a similar story. [ChimeraX](https:\u002F\u002Fwww.rbvi.ucsf.edu\u002Fchimerax\u002F) has a real `movie` command: a mature, scripted renderer with a long track record, and the right call for an animation that has to come out byte-identical on every run. Nanome's answer to motion is a live session somebody can interrupt, which suits a review and doesn't suit a rendered deliverable.\n\nThe pattern that seems to hold: the scene is where the thinking happens, and the flat artifact is what gets filed. Write-ups of projects that ran this way are at [nanome.ai\u002Fcase-studies](https:\u002F\u002Fnanome.ai\u002Fcase-studies).\n\n## FAQ\n\n**How do you present a protein structure to a team?**\nThe presenter loads the structure once, saves a scene for each point the review has to make, and sets each scene's point of view so it opens framed the way the argument needs. Sharing the workspace puts those scenes in front of colleagues in a browser tab or a headset. During the meeting, Spotlight Mode makes the presenter's perspective the shared one, and colored dots show which scene each person is on. Afterwards the scenes stay where they are, so anyone who missed it walks the same order at their own desk.\n\n**What is a molecular scene?**\nA saved 3D view of a structure inside a Nanome workspace. It holds its own camera angle (the point of view, added in v2.1.1), its own representations and its own labels, so scene 2 can show a surface and scene 3 can strip back to sticks on the same molecule. Scenes reorder by dragging, which makes the running order of a talk something you can rearrange 5 minutes before it starts. They shipped with Nanome 2.0, and [visualizing proteins in VR](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fhow-to-visualize-proteins-in-vr) covers getting a structure in there in the first place.\n\n**Can I still get a PowerPoint slide out of it?**\nYes. MARA generates a PowerPoint slide from a scene, so the deck gets made and the 3D is what it came from. The scene stays live in the workspace, which means a question about a different angle is answered by opening it rather than by rebuilding the figure.\n\n**Does the audience need a headset to view the presentation?**\nNo. Shared workspaces open in the browser web app, so a colleague needs a link and a laptop. A headset adds scale and depth for whoever has one, and Nanome runs on [Meta Quest, Apple Vision Pro, HTC Vive Focus 3 and Pico Neo](https:\u002F\u002Fnanome.ai\u002Fsetup). Mixed audiences are the normal case, with some people in a headset and some in a browser tab, moving through the same scenes.\n","2026-08-27T18:19:49.428Z","2026-09-01T01:00:05.584Z","2026-09-01T00:33:45.831Z","2026-08-31","A molecular presentation shows a 3D structure to the people who act on it. How Nanome builds one from scenes, and when a slide is still the right call.","molecular presentation, how to present molecular structures, molecular presentation software, molecular scenes, present a protein structure, 3D molecular presentation","what-a-modern-molecular-presentation-looks-like",{"id":345,"attributes":346},85,{"title":347,"content":348,"createdAt":349,"updatedAt":350,"publishedAt":351,"date":340,"description":352,"keywords":353,"slug":354,"category":17},"How to analyze protein-ligand interactions","To analyze protein-ligand interactions, load the complex into Nanome, turn on interaction lines, and read the hydrogen bonds, hydrophobic contacts, and pi-stacking directly on the 3D structure. Nanome is a collaborative molecular visualization and drug discovery platform that [runs in a browser, on Windows desktop, and in XR headsets](https:\u002F\u002Fnanome.ai\u002Fsetup), and its [AI copilot MARA](https:\u002F\u002Fnanome.ai\u002Fmara) can compute a full interactions report for you. You get the geometry in front of you and the numbers alongside it.\n\nThe short version: you're looking at how a small molecule sits in a binding pocket, and which specific atoms hold it there.\n\n## What to look for\n\n![A researcher wearing an ultra-thin VR headset studies a solid protein surface model with a small ligand visible in its binding pocket held close to the chest](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_to_analyze_protein_ligand_interactions_image_1_v4_5e6675b0b7.png)\n\n\n\nAn interaction analysis comes down to a handful of contact types. Each one tells you something different about why the ligand binds and how tightly.\n\n\u003Ctable class=\"table\">\n  \u003Cthead>\n    \u003Ctr>\u003Cth>Interaction type\u003C\u002Fth>\u003Cth>What it means\u003C\u002Fth>\u003Cth>Why it matters\u003C\u002Fth>\u003C\u002Ftr>\n  \u003C\u002Fthead>\n  \u003Ctbody>\n    \u003Ctr>\u003Ctd>Hydrogen bonds\u003C\u002Ftd>\u003Ctd>A donor and an acceptor atom sharing a hydrogen, usually 2.5 to 3.5 angstroms apart\u003C\u002Ftd>\u003Ctd>The main directional anchor. Miss one and affinity can drop hard.\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Hydrophobic contacts\u003C\u002Ftd>\u003Ctd>Nonpolar surfaces of the ligand tucked against nonpolar residues\u003C\u002Ftd>\u003Ctd>Cheap, forgiving binding energy. Fills the pocket.\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Pi-stacking\u003C\u002Ftd>\u003Ctd>Aromatic rings stacking face-to-face or edge-to-face\u003C\u002Ftd>\u003Ctd>Common with aromatic drugs against Phe, Tyr, Trp, His\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Salt bridges\u003C\u002Ftd>\u003Ctd>A charged group on the ligand paired with an oppositely charged residue\u003C\u002Ftd>\u003Ctd>Strong, but sensitive to pH and solvent\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Steric clashes\u003C\u002Ftd>\u003Ctd>Atoms sitting closer than their radii allow\u003C\u002Ftd>\u003Ctd>A red flag. Often means a bad pose or a strain you'll pay for.\u003C\u002Ftd>\u003C\u002Ftr>\n  \u003C\u002Ftbody>\n\u003C\u002Ftable>\n\nHydrogen bonds are usually the first check, and they're also the easiest to miss if the file arrived without hydrogens in it.\n\n## How to see hydrogen bonds in a protein-ligand complex\n\nHere's the workflow in Nanome, start to finish.\n\n1. **Load the complex.** Open a PDB or SDF file, or fetch a structure by ID from RCSB PDB, DrugBank or PubChem without leaving the app. If the protein and the ligand sit in separate files, load both. The wider tour of getting a structure on screen is in [how to view PDB files in 3D](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fhow-to-view-pdb-files-in-3d-and-the-tools-that-do-it-well).\n2. **Find the ligand.** Select the small molecule and zoom to it. The binding pocket is whatever wraps around it.\n3. **Add hydrogens.** Hydrogen bonds need hydrogens to be defined. Most crystal structures arrive without them, so protonate first and the donors and acceptors become real atoms rather than inferences.\n4. **Turn on interactions.** [Switch on interaction lines](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_v2\u002Ftoolspanel). Nanome draws hydrogen bonds, hydrophobic contacts, and pi-stacking as colored dashes between the ligand and nearby residues, right on the structure.\n5. **Inspect in 3D.** Rotate the pocket. In a browser you do this with the mouse; in a headset you reach out and grab it. Seeing a hydrogen bond from the side tells you whether the angle is any good, which a head-on view flattens away.\n6. **Label the residues.** Show residue names on the contacts, so you can tell whether Asp189 is holding that amine or whether the contact runs to a backbone carbonyl you can't act on.\n\nThat's the visual pass. The quantitative side goes to MARA.\n\n## Letting MARA compute the report\n\n![A flat vector diagram showing a speech bubble on the left connected by an arrow to a structured analysis report on the right, representing plain-English input driving a computational output.](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_to_analyze_protein_ligand_interactions_image_2_4bd07f3254.png)\n\n\n\nThe same job goes to MARA in one sentence. Ask Nanome's copilot to analyze the interactions in your complex and it runs the tools, then hands back a report: the hydrogen bonds with their distances, the hydrophobic contacts, the pi-stacking pairs, and the residues doing the work.\n\nThe report names every tool it called, the inputs it passed in, and what each one returned, so a reviewer can audit the run rather than take a number on faith. Interaction analysis is one small corner of what MARA reaches. The same copilot fires [docking, electrostatics with APBS, ADMET prediction, and co-folding with AlphaFold 3 or Boltz-2](https:\u002F\u002Fnanome.ai\u002Fagents), out of [300+ integrated tools across 26 categories](https:\u002F\u002Fnanome.ai\u002Fintegrations).\n\nNumbers and geometry sit together. The distances come off the report; the angle that produced each one comes off the structure. Handing that result to colleagues who weren't in the session is a separate job, and [what a modern molecular presentation looks like](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fwhat-a-modern-molecular-presentation-looks-like) covers it.\n\n## Where Nanome fits next to other tools\n\n![Two colleagues sit together pointing at a protein ribbon structure on a large display, collaborating on a molecular analysis.](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_to_analyze_protein_ligand_interactions_image_3_6b5e516352.png)\n\n\n\n[PyMOL](https:\u002F\u002Fpymol.org) and [UCSF ChimeraX](https:\u002F\u002Fwww.rbvi.ucsf.edu\u002Fchimerax\u002F) draw contacts well and are the desktop standard, mostly single-user, driven by a GUI or a script. [Schrödinger Maestro](https:\u002F\u002Fwww.schrodinger.com\u002Fplatform\u002Fproducts\u002Fmaestro\u002F) and [MOE](https:\u002F\u002Fwww.chemcomp.com) go deeper again, with contact analysis sitting inside a full comp-chem suite. Nanome's angle on the same pocket is a session several people join at once, native immersive 3D, and a copilot that runs the analysis from a plain-English request. It integrates with several of these suites ([Schrödinger LiveDesign](https:\u002F\u002Fnanome.ai\u002Fintegrations), MOE, Cresset Flare, [OpenEye](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fview-your-openeyecadence-data-with-fresh-eyes-using-nanome-xr)) rather than asking anyone to give them up.\n\nFiles move over without a conversion step. PyMOL `.pse` sessions and `.moe` files open for viewing, Maestro `.mae` and `.maegz` arrive through the LiveDesign gadget, and the ordinary structure files those suites write out load the same way. Those session formats are import only; molecules come back out as PDB, SDF or SMILES, one frame at a time. [More on supported formats](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n\nA binding pocket is a 3D object, and turning one over picks up what a single render angle buries. A hydrogen bond measuring 2.9 angstroms in a table can still sit at a useless angle. A clash tucked behind a side chain shows up the moment the structure turns. With two people inside the same structure, the argument about a pose happens on the atoms themselves, which is the working pattern [collaborative drug discovery software for remote teams](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fcollaborative-drug-discovery-software-for-remote-teams) goes through. Groups have written up how they split that work at [nanome.ai\u002Fcase-studies](https:\u002F\u002Fnanome.ai\u002Fcase-studies).\n\nFor a batch pass over thousands of complexes where no single one gets a look, a headless PyMOL script or a MOE pipeline does the job with fewer moving parts. Nanome fits the other case: one pocket worth understanding, a group that needs to see it together, or a question that's quicker to ask in words than to script.\n\n## FAQ\n\n**How do I see hydrogen bonds in a protein-ligand complex?**\nLoad the complex in Nanome, protonate it if the file arrived without hydrogens, then switch on interaction lines. Hydrogen bonds appear as colored dashes between donor and acceptor atoms. Rotating the pocket in 3D shows the angle as well as the distance, and a bad angle costs you affinity even when the distance looks fine.\n\n**Do I need a headset to analyze interactions?**\nNo. The browser build runs with no headset at all, and there's a Windows desktop app. Headsets (Apple Vision Pro, HTC Vive Focus 3, Meta Quest, Pico Neo) add depth perception, which helps when a pocket is crowded, and the analysis itself is identical either way.\n\n**What files does Nanome load for this?**\nStructures come in as PDB and mmCIF, SDF, MOL and MOL2, XYZ, PQR, SMILES, and AutoDock `.pdbqt` (converted to PDB on load, with charges dropped, new in 2.6.0). Session files load for viewing: PyMOL `.pse`, `.moe`, and Maestro `.mae` \u002F `.maegz`. An electrostatic map (`.dx`) overlays onto a structure already open. Export runs the other way as PDB, SDF or SMILES, single frame. [More on supported formats](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n\n**Does MARA report interaction distances?**\nYes. Ask it to analyze the interactions and it returns hydrogen bond distances, the hydrophobic contacts, and the pi-stacking pairs, along with the tools it called and what each one produced.\n","2026-07-15T01:23:50.401Z","2026-09-01T00:59:59.055Z","2026-09-01T00:33:37.261Z","How to analyze protein-ligand interactions in Nanome: load a complex, show hydrogen bonds and contacts in 3D, and get a MARA report.","how to analyze protein-ligand interactions, how to see hydrogen bonds in a protein-ligand complex, protein-ligand interactions, hydrogen bonds, hydrophobic contacts, pi-stacking, Nanome, MARA","how-to-analyze-protein-ligand-interactions",{"id":356,"attributes":357},83,{"title":358,"content":359,"createdAt":360,"updatedAt":361,"publishedAt":362,"date":340,"description":363,"keywords":364,"slug":365,"category":17},"Collaborative drug discovery software for remote teams","Nanome is collaborative drug discovery software built for remote teams. It runs real-time shared sessions where a chemist in a headset and a colleague on the web app look at the same molecule at the same time, across sites, and reason about it together. It's a molecular visualization and drug discovery platform that spans [XR headsets, Windows desktop, and a browser](https:\u002F\u002Fnanome.ai\u002Fsetup). Nanome's AI copilot, [MARA](https:\u002F\u002Fnanome.ai\u002Fmara), runs the analysis tools inside that same session.\n\nThe problem with remote structural biology is that a protein is a 3D object, and a screen-share flattens it. One person drives, everyone else watches a video of a mouse. Nanome puts the whole group inside one shared 3D room instead, so pointing at a residue means actually pointing at it.\n\n## What a shared session looks like\n\n![A researcher wearing an ultra-thin VR headset studies a ribbon-cartoon protein structure held close to the chest](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fcollaborative_drug_discovery_software_for_remote_teams_image_1_v4s_4a7ea5d283.png)\n\n\n\nA docked pose needs a second opinion. A medicinal chemist in San Diego puts on a Meta Quest and steps into the binding pocket at room scale. A structural biologist in Boston opens the same session in a browser on a laptop, no headset needed. They're in one workspace, looking at one molecule.\n\nThe chemist grabs the ligand with their hands and rotates it. The biologist sees it move in real time and points at a clash near a backbone carbonyl. Both know which atom, because they're pointing at the same object in the same space.\n\nThat's the whole idea. Presence in place of a screen-share. Everyone gets depth, everyone gets scale, and nobody has to say \"no, the other one, up and to the left.\"\n\n## Collaboration features for remote teams\n\n\u003Ctable class=\"table\">\n  \u003Cthead>\n    \u003Ctr>\u003Cth>Feature\u003C\u002Fth>\u003Cth>What it does for a remote team\u003C\u002Fth>\u003C\u002Ftr>\n  \u003C\u002Fthead>\n  \u003Ctbody>\n    \u003Ctr>\u003Ctd>Real-time shared sessions\u003C\u002Ftd>\u003Ctd>Several people join one workspace and see every change live, no screen-share\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Cross-device presence\u003C\u002Ftd>\u003Ctd>A headset user and a web user share the same molecule at the same time\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Web app, no headset\u003C\u002Ftd>\u003Ctd>Anyone can join from a browser on a laptop, so headset access never blocks a review\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Native immersive 3D\u003C\u002Ftd>\u003Ctd>Depth and scale are seen directly, so a whole team reads the structure the same way\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Hand-based manipulation\u003C\u002Ftd>\u003Ctd>Grab, rotate, and resize a structure so collaborators follow the reasoning as it happens\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Shared file loading\u003C\u002Ftd>\u003Ctd>Fetch a structure by code from RCSB PDB, UniProt, ChEMBL, PubChem or DrugBank, or drop in a file, and it lands in the live session\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>MARA in the room\u003C\u002Ftd>\u003Ctd>Ask in plain English, run a tool, and the result lands in the scene everyone is looking at\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Behind-the-firewall deploy\u003C\u002Ftd>\u003Ctd>Single-tenant cloud or on-prem, so IP stays inside the building\u003C\u002Ftd>\u003C\u002Ftr>\n  \u003C\u002Ftbody>\n\u003C\u002Ftable>\n\n## Web and XR in the same room\n\nThe web app is what makes this practical for a real team. Headsets are rarely one per person, and a review can't wait for a device to free up. A collaborator joins from a browser, gets the same shared structure, and takes part fully.\n\nThe headsets are Meta Quest, Pico Neo, Apple Vision Pro and HTC Vive Focus 3. Windows desktop and the browser cover everyone else. A single session mixes them freely: two people in headsets, three on laptops, one molecule, all in sync.\n\nStructures arrive by code from RCSB PDB, UniProt, ChEMBL, PubChem or DrugBank, or straight off someone's disk, so a session can open with the real thing in front of the group in seconds. For the wider tour of what opens a structure in 3D, we went through the options in [how to view PDB files in 3D](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fhow-to-view-pdb-files-in-3d-and-the-tools-that-do-it-well).\n\n## Running real tools together\n\nLooking at a structure is half of a review. The other half is the calculation that says whether a change helps.\n\nMARA handles that half. You ask in plain English, MARA runs the tool, and the result returns into the same shared scene. It reaches [300+ integrated scientific tools](https:\u002F\u002Fnanome.ai\u002Fintegrations). Every run is recorded with its inputs and its outputs, so a colleague who joins the session late can trace how a number was produced.\n\nA few of the things it runs:\n\n- Docking with Smina or DiffDock-L, and co-folding\n- Structure prediction with [AlphaFold 3](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41586-024-07487-w), [Boltz-2](https:\u002F\u002Fgithub.com\u002Fjwohlwend\u002Fboltz) and Chai-1\n- Electrostatics through APBS\n- ADMET and toxicity prediction\n- Sequence design with ProteinMPNN, and ANARCI for antibody numbering and CDR loops\n- De novo binder design with RFdiffusion3 (beta)\n- Cheminformatics\n\nA remote team can dock a compound into the pocket they're standing in and argue about the result on the spot, with nobody exporting a file and emailing it around. From there the group is reading contacts together, and [how to analyze protein-ligand interactions](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fhow-to-analyze-protein-ligand-interactions) covers what a careful pass looks like.\n\n## From calculation to shared decision\n\n![Two colleagues in a lounge area review a molecular dynamics trajectory together on a shared display, one gesturing toward the structure.](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fcollaborative_drug_discovery_software_for_remote_teams_image_2_7d3f5e7b31.png)\n\n\n\nA result moves a program only once everyone reads it the same way, and that is where distributed groups stall. Novartis ran its COVID-19 drug discovery discussions inside Nanome in 2020, with colleagues in the United States and Switzerland joining from their homes. The molecule was in the room even though the team wasn't.\n\nTime zones do more damage to a review than distance does. A colleague 8 hours out can open [the sequence of views you saved](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fwhat-a-modern-molecular-presentation-looks-like) and walk it at their own desk, rather than waiting for an hour that suits both clocks.\n\n## Enterprise and behind-the-firewall\n\n![A flat vector diagram showing data, molecules, and computation contained entirely within a secure building perimeter, nothing crossing outside.](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fcollaborative_drug_discovery_software_for_remote_teams_image_3_d588a2afdc.png)\n\n\n\nDrug discovery IP can't leave the building, and remote collaboration usually means data crossing sites. Nanome's enterprise deployments, MARA included, sit [behind your own firewall](https:\u002F\u002Fnanome.ai\u002Fmara-enterprise) as single-tenant cloud or on-prem hardware. A REST API and MCP servers are open for internal tooling, and there's a Nanome Claude Code Skill for teams that work that way.\n\nNanome also plugs into pipelines a group already runs: [Schrödinger LiveDesign](https:\u002F\u002Fnanome.ai\u002Fintegrations), Cresset Flare, [OpenEye \u002F Cadence](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fview-your-openeyecadence-data-with-fresh-eyes-using-nanome-xr), [CDD Vault](https:\u002F\u002Fwww.collaborativedrug.com), [KNIME](https:\u002F\u002Fwww.knime.com) and Jupyter. When the pipeline lives in one of those, Nanome becomes the front end where a distributed team reviews its output together.\n\n## How Nanome compares for remote collaboration\n\n\u003Ctable class=\"table\">\n  \u003Cthead>\n    \u003Ctr>\u003Cth>Option\u003C\u002Fth>\u003Cth>Strength\u003C\u002Fth>\u003Cth>What a shared session adds\u003C\u002Fth>\u003C\u002Ftr>\n  \u003C\u002Fthead>\n  \u003Ctbody>\n    \u003Ctr>\u003Ctd>\u003Ca href=\"https:\u002F\u002Fpymol.org\">PyMOL\u003C\u002Fa>, \u003Ca href=\"https:\u002F\u002Fwww.rbvi.ucsf.edu\u002Fchimerax\u002F\">ChimeraX\u003C\u002Fa>, \u003Ca href=\"https:\u002F\u002Fwww.ks.uiuc.edu\u002FResearch\u002Fvmd\u002F\">VMD\u003C\u002Fa>\u003C\u002Ftd>\u003Ctd>Desktop viewers, mostly single-user, script or GUI driven\u003C\u002Ftd>\u003Ctd>Real-time multiplayer rooms that span web and XR\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>\u003Ca href=\"https:\u002F\u002Fwww.schrodinger.com\u002Fplatform\u002Fproducts\u002Fmaestro\u002F\">Schrödinger Maestro\u003C\u002Fa>, \u003Ca href=\"https:\u002F\u002Fwww.chemcomp.com\">MOE\u003C\u002Fa>, \u003Ca href=\"https:\u002F\u002Fwww.3ds.com\u002Fproducts\u002Fbiovia\u002Fdiscovery-studio\">BIOVIA Discovery Studio\u003C\u002Fa>\u003C\u002Ftd>\u003Ctd>Full desktop comp-chem suites\u003C\u002Ftd>\u003Ctd>Nanome integrates with several of them, then puts the output in a room the team can walk into\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Screen-share over a video call\u003C\u002Ftd>\u003Ctd>Quick way to show a structure to one room\u003C\u002Ftd>\u003Ctd>Everyone holds their own view of one live molecule\u003C\u002Ftd>\u003C\u002Ftr>\n  \u003C\u002Ftbody>\n\u003C\u002Ftable>\n\n## Where a desktop viewer still wins\n\nFor solo work, like scripting a figure for a paper, a desktop viewer such as PyMOL or ChimeraX is quicker. When the team needs to gather around the same structure, Nanome opens PyMOL `.pse` sessions directly, along with the everyday structure files: PDB, mmCIF, SDF, XYZ, PQR and MOL\u002FMOL2.\n\nIf a calculation lives inside a [Schrödinger](https:\u002F\u002Fwww.schrodinger.com) or [MOE](https:\u002F\u002Fwww.chemcomp.com) workflow, that workflow can stay where it is. Nanome imports Maestro `.mae` and `.maegz` files, the format LiveDesign also ingests, plus MOE `.moe` files, so the output opens in a shared session for the group to gather around.\n\nGroups already work across sites and clocks this way, and their stories are written up at [nanome.ai\u002Fcase-studies](https:\u002F\u002Fnanome.ai\u002Fcase-studies).\n\n## FAQ\n\n**What is collaborative drug discovery software for remote teams?**\nSoftware that lets a distributed team view and work on the same molecular structure together in real time. Nanome does it with shared sessions spanning XR headsets, Windows desktop and a browser, so a headset user and a web user hold one live molecule between them.\n\n**What is the best software for remote structural biology collaboration?**\nNanome is built for it. Several people join one shared 3D session across web and XR, point at the same residues, and run real tools through MARA, with every change synced across sites.\n\n**Do remote collaborators need a headset?**\nNo. There's a browser web app and a Windows desktop app, so a colleague can join from a laptop while teammates work in full immersion on headsets.\n\n**What file formats can a shared session open?**\nStructures import as PDB (`.pdb`, `.ent`), mmCIF (`.cif`, `.mmcif`), SDF, MOL and MOL2, SMILES, XYZ, PQR, and PDBQT (converted to PDB, with charges dropped). Session and vendor files import too: PyMOL `.pse`, Maestro `.mae` and `.maegz`, and MOE `.moe`. An electrostatic map (`.dx`) overlays onto a structure that is already loaded. Export is PDB, SDF or SMILES, single frame, which makes mmCIF, MAE, MOE and PSE import-only. Full detail sits in [the format docs](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n\n**Can a remote team keep its data behind the firewall?**\nYes. Enterprise deployments, MARA included, run as single-tenant cloud or on-prem, so structures and results stay inside your environment.\n","2026-07-15T01:23:50.201Z","2026-09-01T00:59:57.950Z","2026-09-01T00:33:35.573Z","Collaborative drug discovery software for remote teams: Nanome runs real-time shared molecular sessions across web and XR, behind your firewall.","collaborative drug discovery software, software for remote structural biology collaboration, remote drug discovery, real-time molecular collaboration, Nanome, MARA, XR drug discovery","collaborative-drug-discovery-software-for-remote-teams",{"id":367,"attributes":368},103,{"title":369,"content":370,"createdAt":371,"updatedAt":372,"publishedAt":373,"date":340,"description":374,"keywords":375,"slug":376,"category":17},"VR for medicinal chemists","Nanome is a collaborative molecular visualization and drug discovery platform that runs on [XR headsets, Windows desktop, and a browser web app](https:\u002F\u002Fnanome.ai\u002Fsetup). For a medicinal chemist, VR turns a flat picture of a binding pocket into a real object you can walk around, reach into, and discuss with your team standing next to you. An AI copilot called [MARA](https:\u002F\u002Fnanome.ai\u002Fmara) runs the tools ([docking](https:\u002F\u002Fdocs.nanome.ai\u002Fmara\u002Ffeatures), property calculations, and more) from plain-English requests, so the analysis happens right where you are looking at the molecule.\n\nMost of the day-to-day still runs through [Maestro](https:\u002F\u002Fwww.schrodinger.com\u002Fplatform\u002Fproducts\u002Fmaestro\u002F), [MOE](https:\u002F\u002Fwww.chemcomp.com), or a 2D sketcher. VR comes in when the 3D is the question and a screenshot stops carrying the answer.\n\n## What a medicinal chemist gets out of it\n\n![A medicinal chemist wearing an ultra-thin VR headset turns a protein-ligand surface model floating just in front of her chest in a bright open studio.](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fvr_for_medicinal_chemists_image_1_r3_7ae9957e5e.png)\n\n\n\nA 2D depiction flattens the geometry the pocket cares about. A methyl that looks harmless on paper can bump a backbone carbonyl. A hydrogen bond drawn as a dash can sit 40 degrees off the angle it needs.\n\nIn VR the ligand sits in the pocket at true scale, with both hands free to turn it and lean in. Depth is real, so a clash reads as a clash, and pointing at an atom means pointing at that atom rather than calling it the one near the top left.\n\nReading those contacts in detail, hydrogen bonds, hydrophobic contacts, pi stacking, is a job of its own, and [how to analyze protein-ligand interactions](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fhow-to-analyze-protein-ligand-interactions) works through it.\n\nDesign review changes shape too. Several chemists stand around one pocket and mark up the same ligand while they talk, since Nanome hosts real-time multiplayer sessions, so San Diego and Basel can hold the same molecule at once. [Collaborative drug discovery software for remote teams](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fcollaborative-drug-discovery-software-for-remote-teams) covers how those sessions run across sites.\n\nNimbus Therapeutics ran the AMPKβ2 enzyme through exactly that kind of review. The team had a selectivity strategy they considered settled, and watching the protein in VR exposed a better synthetic vector. They changed course, and the compounds came out more active on the target.\n\n## What VR adds for med chem\n\n\u003Ctable class=\"table\">\n  \u003Cthead>\n    \u003Ctr>\u003Cth>Task\u003C\u002Fth>\u003Cth>On a flat screen\u003C\u002Fth>\u003Cth>With Nanome in VR\u003C\u002Fth>\u003C\u002Ftr>\n  \u003C\u002Fthead>\n  \u003Ctbody>\n    \u003Ctr>\u003Ctd>SAR discussion\u003C\u002Ftd>\u003Ctd>Rotating a static image, atoms described in words\u003C\u002Ftd>\u003Ctd>Everyone around one pocket at scale, pointing at the exact atom and H-bond\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Design review\u003C\u002Ftd>\u003Ctd>A screen-share with one driver\u003C\u002Ftd>\u003Ctd>A shared 3D session, several people marking the same ligand at once\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Seeing a ligand in context\u003C\u002Ftd>\u003Ctd>Depth inferred from a 2D projection\u003C\u002Ftd>\u003Ctd>True depth, real clashes, a pocket shape you can lean into\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Running docking or ADMET\u003C\u002Ftd>\u003Ctd>Switch apps, export files, wait, re-import\u003C\u002Ftd>\u003Ctd>MARA runs it inside the session, results land on the molecule you are holding\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Handing off to a colleague\u003C\u002Ftd>\u003Ctd>Send a file and a paragraph of setup notes\u003C\u002Ftd>\u003Ctd>Send the workspace with your saved views, so they open on the angle you picked\u003C\u002Ftd>\u003C\u002Ftr>\n  \u003C\u002Ftbody>\n\u003C\u002Ftable>\n\nThat last row is the one that travels furthest. A saved view carries the argument a file cannot: the colleague opens the workspace already framed on the pocket, at the orientation the discussion settled on, and [what a modern molecular presentation looks like](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fwhat-a-modern-molecular-presentation-looks-like) builds the rest of the walkthrough from there.\n\n## MARA runs the tools without leaving the molecule\n\n![A plain-language request arrow flows into a grid of four scientific tool icons, showing how MARA routes a spoken command to the right analysis.](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fvr_for_medicinal_chemists_image_2_fa5601b219.png)\n\n\n\nThe request goes in as plain English. MARA picks the tool, runs it, and reports back what fired, what went in, and what came out. [300+ tools across 26 categories](https:\u002F\u002Fnanome.ai\u002Fintegrations) sit behind that one request. Inside a headset MARA takes voice commands, so a run starts while both hands stay on the structure.\n\nIn a med chem workflow that looks like:\n\n- Docking a fresh analog into the pocket in front of you, on Smina or DiffDock-L, then comparing poses side by side.\n- ADMET predictions across a short series, before any of it goes into synthesis.\n- APBS electrostatics, to see where the pocket carries charge before a substituent gets picked.\n- A co-folded or predicted model from [AlphaFold 3](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41586-024-07487-w) or [Boltz-2](https:\u002F\u002Fgithub.com\u002Fjwohlwend\u002Fboltz) when no crystal structure exists yet.\n\nStructures arrive from RCSB PDB, PubChem, ChEMBL, DrugBank and UniProt.\n\n## Where Nanome sits in a chemistry stack\n\n\u003Cdiv class=\"logo-row\" style=\"display:flex;flex-wrap:wrap;align-items:center;justify-content:center;gap:1.75rem;background:#fff;border-radius:12px;padding:1.25rem 1.5rem;margin:2rem 0\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fvr_for_medicinal_chemists_image_5_51bc98da2d.png\" alt='MOE' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fvr_for_medicinal_chemists_image_7_9fd8ce0166.png\" alt='PyMOL' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fvr_for_medicinal_chemists_image_9_116dc74cb2.png\" alt='OpenEye' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003C\u002Fdiv>\n\n\n\n![Two colleagues in a research lounge discuss a protein ribbon structure shown on a wall display, representing how Nanome sits alongside existing computational tools.](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fvr_for_medicinal_chemists_image_3_d5b412b661.png)\n\n\n\nNanome plugs into a chemistry stack rather than standing in for one. [CDD Vault](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fcollaborative-drug-discovery-and-nanome-partnership-announcement), [Schrödinger LiveDesign](https:\u002F\u002Fnanome.ai\u002Fintegrations), Cresset Flare, [OpenEye](https:\u002F\u002Fwww.eyesopen.com)\u002FCadence, [KNIME](https:\u002F\u002Fwww.knime.com) and Jupyter all have a path in.\n\nThe division of work is easy to state. Maestro, MOE and [Discovery Studio](https:\u002F\u002Fwww.3ds.com\u002Fproducts\u002Fbiovia\u002Fdiscovery-studio) hold the heavy comp-chem jobs: large batches, parameter-heavy runs, anything scripted. [PyMOL](https:\u002F\u002Fpymol.org) and [ChimeraX](https:\u002F\u002Fwww.rbvi.ucsf.edu\u002Fchimerax\u002F) hold publication figures and precise single-user editing at a desk. Nanome takes the part where the 3D and the conversation happen at the same time: SAR around a real pocket, a design review the whole team stands inside, and MARA running docking or property tools on the molecule in your hands.\n\nGetting a structure across takes no conversion step. Nanome imports Maestro `.mae` and `.maegz` files, the same pair LiveDesign hands over, opens MOE `.moe` files, and reads PyMOL `.pse` sessions, though a session carrying QM\u002FMM link atoms can trip that parser. All 3 of those come in one direction only. The standard structure formats load as well: PDB, mmCIF, SDF, MOL\u002FMOL2, XYZ, PQR, SMILES and PDBQT. What Nanome writes back out is narrower, PDB, SDF or SMILES, one frame at a time. [More on supported formats](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n\nA scripted virtual screen across a million compounds has no reason to run in a headset. Getting 4 chemists to agree on the next 3 analogs while they all stand around one pocket is the case VR was built for. Write-ups of projects that ran that way sit at [nanome.ai\u002Fcase-studies](https:\u002F\u002Fnanome.ai\u002Fcase-studies).\n\n## FAQ\n\n**What is VR for medicinal chemistry?**\nImmersive 3D used to inspect and design molecules at true scale. In Nanome that means a ligand viewed inside its binding pocket, turned with both hands, with clashes and H-bond geometry visible that a 2D drawing flattens, and analysis run on it through the MARA copilot.\n\n**How do medicinal chemists use VR day to day?**\nMostly SAR discussions and design reviews. A team shares one 3D scene, points at specific atoms, marks up ideas together, and asks MARA to dock an analog or return ADMET numbers without leaving the session. Saved views then carry the outcome to whoever missed it.\n\n**Do I need a headset?**\nNo. Nanome runs on Meta Quest and HTC Vive Focus 3 headsets, on Pico Neo and on Apple Vision Pro. It also runs as a Windows desktop app, and in a browser web app that needs nothing but a laptop. Reviews often start in the web app and move into a headset once the 3D is carrying the argument.\n\n**Does it replace Maestro or MOE?**\nNo. It sits alongside them and connects to Schrödinger LiveDesign, Cresset Flare, OpenEye\u002FCadence and CDD Vault. The suites keep the scripted and batch work; Nanome takes immersive review and collaboration.\n\n**What file formats does Nanome read?**\nImport covers PDB, mmCIF, SDF, MOL\u002FMOL2, XYZ, PQR, SMILES and PDBQT, plus the proprietary and session files: Maestro `.mae` and `.maegz`, MOE `.moe`, and PyMOL `.pse`. Those last four are import-only. Export is PDB, SDF or SMILES, a single frame. The tier-by-tier list lives in the [supported file formats](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats) docs.\n","2026-07-15T01:23:52.345Z","2026-09-01T01:00:04.406Z","2026-09-01T00:33:44.090Z","VR for medicinal chemistry, explained. How medicinal chemists use Nanome to review SAR, ligands, and pockets in immersive 3D.","VR for medicinal chemistry, how medicinal chemists use VR, medicinal chemistry VR, SAR in VR, ligand design VR, Nanome, MARA","vr-for-medicinal-chemists",{"id":378,"attributes":379},90,{"title":380,"content":381,"createdAt":382,"updatedAt":383,"publishedAt":384,"date":340,"description":385,"keywords":386,"slug":387,"category":17},"Molecular visualization for structural biologists","Molecular visualization for structural biology means loading a solved structure, inspecting it residue by residue, and checking the parts that decide whether a model holds up: interfaces, multimers, and alignments against related structures. Nanome is a collaborative molecular visualization and drug discovery platform that does this in immersive 3D, in a browser web app and in XR headsets. Inside it, an [AI copilot called MARA](https:\u002F\u002Fnanome.ai\u002Fmara) runs the structural analysis. It loads PDB and mmCIF structures and pairs with the desktop tools structural biologists already run, [PyMOL](https:\u002F\u002Fpymol.org), [UCSF ChimeraX](https:\u002F\u002Fwww.rbvi.ucsf.edu\u002Fchimerax\u002F), and [Coot](https:\u002F\u002Fwww2.mrc-lmb.cam.ac.uk\u002Fpersonal\u002Fpemsley\u002Fcoot\u002F).\n\nThree questions come up on nearly every structure. Does this chain really contact that one? Is the multimer biological or a crystal artifact? How does the new model line up against the reference?\n\nA viewer has to make all three legible from the coordinates alone.\n\n## What structural biologists need from a viewer\n\n\u003Cdiv class=\"logo-row\" style=\"display:flex;flex-wrap:wrap;align-items:center;justify-content:center;gap:1.75rem;background:#fff;border-radius:12px;padding:1.25rem 1.5rem;margin:2rem 0\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fmolecular_visualization_for_structural_biologists_image_4_c9aa13248f.png\" alt='PyMOL' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003C\u002Fdiv>\n\n\n\n![A researcher wearing an ultra-thin VR headset studies a ribbon-cartoon protein structure held close to the chest](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fmolecular_visualization_for_structural_biologists_image_1_v4s_8be253ba91.png)\n\n\n\nMost of the work happens close to the coordinates, chain by chain, with the attention landing on the places where a build tends to go wrong.\n\nInterfaces carry the story. One contact modelled wrong can change what the whole complex is supposed to be doing, and the deposited assembly is not always the biological one, so multimer calls get re-argued long after deposition.\n\nA new structure gets most of its meaning by comparison, which is why alignment against earlier models is rarely optional. Checking a build against what the density actually supports stays in Coot, since that is the job Coot was written for.\n\nHere is how the common tools divide the work, and what Nanome adds next to them.\n\n\u003Ctable class=\"table\">\n  \u003Cthead>\n    \u003Ctr>\u003Cth>Tool\u003C\u002Fth>\u003Cth>Its specialty\u003C\u002Fth>\u003Cth>What Nanome adds beside it\u003C\u002Fth>\u003C\u002Ftr>\n  \u003C\u002Fthead>\n  \u003Ctbody>\n    \u003Ctr>\u003Ctd>PyMOL\u003C\u002Ftd>\u003Ctd>Ray-traced figures, scripted rendering, 20+ years of community scripts\u003C\u002Ftd>\u003Ctd>Reads the same PDB and mmCIF, opens a \u003Ccode>.pse\u003C\u002Fcode> session for viewing, and puts several people around one structure at once\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>UCSF ChimeraX\u003C\u002Ftd>\u003Ctd>Density maps, cryo-EM, very large assemblies, deep analysis\u003C\u002Ftd>\u003Ctd>Takes the same coordinates into XR so a group can walk an interface together\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Coot\u003C\u002Ftd>\u003Ctd>Model building and refinement against density\u003C\u002Ftd>\u003Ctd>Picks up downstream, as the shared review once a build is ready to be argued over\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Nanome\u003C\u002Ftd>\u003Ctd>Shared immersive 3D review with an AI copilot\u003C\u002Ftd>\u003Ctd>MARA runs alignment, interface detection, and SASA, naming the tool and the inputs behind every result\u003C\u002Ftd>\u003C\u002Ftr>\n  \u003C\u002Ftbody>\n\u003C\u002Ftable>\n\nEach desktop tool is the best thing going for its own job. Nanome covers the part that sits between them: several people inside one structure at the same time, at true 3D scale, with the analysis one request away.\n\n## How Nanome helps with structural review\n\n![Two colleagues wearing ultra-thin VR headsets examine the same space-filling protein model floating between them](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fmolecular_visualization_for_structural_biologists_image_2_v4_28478c3648.png)\n\n\n\nA monitor flattens a 3D fold onto a plane. Nanome [renders the structure natively in XR on Apple Vision Pro, Meta Quest, Pico Neo, and HTC Vive Focus 3](https:\u002F\u002Fnanome.ai\u002Fsetup). There's also a Windows build and a browser app that runs with no headset at all.\n\nIn a headset a multimer is something to walk around, and an interface is something to put both hands into and open.\n\nSessions are real-time multiplayer. A structural biologist, a crystallographer, and a computational chemist can stand in one model and point at the same loop, which is a different act from describing that loop over a screen share. [Collaborative drug discovery software for remote teams](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fcollaborative-drug-discovery-software-for-remote-teams) covers how those sessions get run across time zones.\n\nThat gap is what the 2019 platform paper set out to close. Kingsley et al., [Development of a virtual reality platform for effective communication of structural data in drug discovery](https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.jmgm.2019.03.010), J. Mol. Graph. Model. 89, written with a Novartis GNF team, opens on a single kinase binding site drawn three different ways in 2D. Each drawing keeps something the other two lose, and the paper's worry is that the medicinal chemist or biologist reading them ends up in the dark about which one to trust.\n\nAn interface claim is a 3D claim, and a figure of an interface fixes one point of view before the reviewer has picked theirs. That is the argument in [what a modern molecular presentation looks like](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fwhat-a-modern-molecular-presentation-looks-like).\n\nThe analysis runs through MARA. Describe the job in plain English and it covers the structural-biology staples: [align a reference structure onto a set of mobile ones, find the interface residues between two chains, and compute solvent accessible surface area (SASA) per residue](https:\u002F\u002Fdocs.nanome.ai\u002Fmara\u002Ffeatures). Every answer names the tool it called and the inputs it used, so a result can be checked rather than taken on trust. [How to analyze protein-ligand interactions](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fhow-to-analyze-protein-ligand-interactions) walks the same machinery over a bound ligand.\n\nStructures arrive from RCSB PDB, PubChem, and DrugBank without anyone leaving the session. From there MARA reaches [300+ integrated scientific tools](https:\u002F\u002Fnanome.ai\u002Fintegrations) across 26 categories: docking with Smina and DiffDock-L, co-folding and structure prediction with [AlphaFold 3](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41586-024-07487-w) and [Boltz-2](https:\u002F\u002Fgithub.com\u002Fjwohlwend\u002Fboltz), electrostatics through APBS, ADMET prediction, ANARCI for antibody numbering and CDR definition, ProteinMPNN for sequence design, RFdiffusion3 (beta) for de novo binders, cheminformatics, and MD trajectory analysis.\n\nTrajectories come across from your simulation engine. Load a `.gro` standalone, or attach `.xtc`, `.trr`, or `.dcd` frames to a model that's already open, matching atom counts. Frame playback carries a 2000-frame cap, and surfaces are off during playback. [More on supported formats](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n\n## Jobs that stay on the desktop\n\n![A researcher at a plain wooden table studies a smooth molecular surface displayed on a large monitor in a quietly lit research office.](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fmolecular_visualization_for_structural_biologists_image_3_72f7dbc723.png)\n\n\n\nModel building and refinement against density belong in Coot, which is what it was written for. A polished static figure for a paper is PyMOL's job. Cryo-EM maps and very large assemblies are ChimeraX territory.\n\nNanome opens the same standard files those tools write: PDB, mmCIF, SDF, MOL2, XYZ, PQR, and PDBQT. It also loads a PyMOL `.pse` session for viewing, though the validated session versions aren't documented and QM\u002FMM link atoms can break the parse, so a test load is worth the minute before a meeting depends on the file. [More on supported formats](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n\nThe practical pattern is a round trip. The structure moves into Nanome when a group needs to look at it together at real scale, then back to the desktop for the refinement, the rendering, and the map work. The [case studies](https:\u002F\u002Fnanome.ai\u002Fcase-studies) show what that review step looked like on live programs.\n\n## FAQ\n\n**What is molecular visualization for structural biology?**\nLoading a solved structure and inspecting the parts that decide model quality: chains, interfaces, multimers, and how the model aligns against related structures. Nanome does that in immersive 3D, reads PDB and mmCIF, and runs alignment, interface detection, and SASA through MARA.\n\n**Which tools do structural biologists use to view structures?**\nThe desktop set is PyMOL for figures and inspection, UCSF ChimeraX for density maps and large assemblies, and Coot for building and refinement against density. Nanome joins that set as the shared 3D review step and as the route into structural analysis through MARA.\n\n**What file formats does Nanome support?**\nImport covers the structure formats a structural biology group already has on disk: PDB (`.pdb`, `.ent`), mmCIF and PDBx (`.cif`, `.mmcif`, `.mcif`, `.bcif`), SDF (`.sdf`, `.sd`, `.mol`), MOL2, SMILES, XYZ, PQR, and PDBQT. Session and vendor files load as well: PyMOL `.pse`, Maestro `.mae` and `.maegz`, and MOE `.moe`. MD trajectories are their own tier, with `.gro` loading standalone and `.xtc`, `.trr`, and `.dcd` attaching to an open model of matching atom count. A `.dx` electrostatic map overlays a model that's already loaded. Writing back out is narrower: PDB, SDF, or SMILES, one frame at a time, which leaves mmCIF, MAE, MOE, and PSE import-only. Electron density maps in CCP4, MRC, or DSN6 aren't supported. [More on supported formats](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n\n**Does Nanome need a headset?**\nNo. There's a browser app and a Windows desktop build. Headset support covers Apple Vision Pro, Meta Quest, Pico Neo, and HTC Vive Focus 3, for the work where true 3D scale is the point.\n","2026-07-15T01:23:50.914Z","2026-09-01T01:00:03.496Z","2026-09-01T00:33:42.380Z","Molecular visualization for structural biology: how Nanome pairs with PyMOL, ChimeraX, and Coot for immersive 3D review of PDB and mmCIF structures.","molecular visualization for structural biology, tools structural biologists use to view structures, PyMOL, UCSF ChimeraX, Coot, PDB, mmCIF, interfaces, multimers, Nanome","molecular-visualization-for-structural-biologists",{"id":389,"attributes":390},102,{"title":391,"content":392,"createdAt":393,"updatedAt":394,"publishedAt":395,"date":340,"description":396,"keywords":397,"slug":398,"category":17},"How to view PDB files in 3D, and the tools that do it well","To view a PDB file in 3D, open it in a molecular viewer like [PyMOL](https:\u002F\u002Fpymol.org), [UCSF ChimeraX](https:\u002F\u002Fwww.rbvi.ucsf.edu\u002Fchimerax\u002F), or [Mol*](https:\u002F\u002Fmolstar.org), then pick a representation (cartoon for the protein, sticks for a ligand). Nanome does this too, opening PDB and SDF files in [a browser web app or in XR](https:\u002F\u002Fnanome.ai\u002Fsetup), and it pulls structures straight from the RCSB PDB by 4-character ID. A PDB file is just a text format holding atomic coordinates, so any of these tools reads it directly.\n\n## The 3-minute how-to\n\n\u003Cdiv class=\"logo-row\" style=\"display:flex;flex-wrap:wrap;align-items:center;justify-content:center;gap:1.75rem;background:#fff;border-radius:12px;padding:1.25rem 1.5rem;margin:2rem 0\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_to_view_pdb_files_in_3d_and_the_tools_that_do_it_well_image_4_59f0136741.png\" alt='PyMOL' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_to_view_pdb_files_in_3d_and_the_tools_that_do_it_well_image_6_47eacf10bc.png\" alt='Mol*' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_to_view_pdb_files_in_3d_and_the_tools_that_do_it_well_image_7_f2501d253e.png\" alt='VMD' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_to_view_pdb_files_in_3d_and_the_tools_that_do_it_well_image_9_0d393ccc6e.png\" alt='Cresset Flare' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003C\u002Fdiv>\n\n\n\n![A researcher wearing an ultra-thin VR headset studies a ribbon-cartoon protein structure held close to the chest](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_to_view_pdb_files_in_3d_and_the_tools_that_do_it_well_image_1_v4s_111d5ebf30.png)\n\n\n\nFour steps take a 4-character PDB ID to a structure that turns on screen.\n\n1. **Getting the file.** [rcsb.org](https:\u002F\u002Fwww.rcsb.org) takes a protein name (say hemoglobin) or an ID like 4HHB and hands back a `.pdb` or a `.cif`. Small molecules from PubChem or DrugBank arrive as SDF instead.\n2. **Opening it.** [molstar.org\u002Fviewer](https:\u002F\u002Fmolstar.org\u002Fviewer) accepts a dragged-in file with nothing installed. PyMOL, ChimeraX and Nanome's web app open the same file.\n3. **Picking a representation.** Protein backbones read best as cartoon or ribbon. Ligands and side chains read best as sticks or ball-and-stick. A surface is what shows the shape of a pocket.\n4. **Getting oriented.** Every viewer here rotates on a mouse drag and zooms on scroll, and coloring by chain or by element separates the parts. After that the questions get specific: which residues line the pocket, what the ligand actually touches, and that's [analyzing protein-ligand interactions](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fhow-to-analyze-protein-ligand-interactions).\n\nThat's the loop. A typical PDB structure opens in any of these without special handling.\n\n## Common viewers, and where Nanome fits\n\n![A flat diagram showing four separate viewer tools converging into one collaborative platform where multiple users share a single structure.](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_to_view_pdb_files_in_3d_and_the_tools_that_do_it_well_image_3_711257863d.png)\n\n\n\n\u003Ctable class=\"table\">\n  \u003Cthead>\n    \u003Ctr>\u003Cth>Viewer\u003C\u002Fth>\u003Cth>Where it's strong\u003C\u002Fth>\u003Cth>How Nanome overlaps\u003C\u002Fth>\u003C\u002Ftr>\n  \u003C\u002Fthead>\n  \u003Ctbody>\n    \u003Ctr>\u003Ctd>PyMOL\u003C\u002Ftd>\u003Ctd>Publication-quality renders, scripting, a deep plugin ecosystem\u003C\u002Ftd>\u003Ctd>Nanome imports PyMOL \u003Ccode>.pse\u003C\u002Fcode> sessions and puts live multi-user review and native XR around the same PDB and SDF coordinates. Traffic runs one way, since there's no \u003Ccode>.pse\u003C\u002Fcode> writer. \u003Ca href=\"https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats\">More on supported formats\u003C\u002Fa>.\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>UCSF ChimeraX\u003C\u002Ftd>\u003Ctd>Analysis, density maps, large complexes on the desktop\u003C\u002Ftd>\u003Ctd>Nanome opens the same standard coordinate files (PDB, mmCIF, SDF, MOL2, XYZ, PQR) and adds a session colleagues can join, with RCSB fetch by ID inside the app. \u003Ca href=\"https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats\">More on supported formats\u003C\u002Fa>.\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Mol*\u003C\u002Ftd>\u003Ctd>Fast in-browser viewing, no install, clean embedding in web pages\u003C\u002Ftd>\u003Ctd>Nanome's web app runs in the browser as well, reads the same PDB, mmCIF and SDF files, and brings headset support and the \u003Ca href=\"https:\u002F\u002Fnanome.ai\u002Fmara\">MARA\u003C\u002Fa> copilot with it. \u003Ca href=\"https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats\">More on supported formats\u003C\u002Fa>.\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>VMD\u003C\u002Ftd>\u003Ctd>Large systems, scripted analysis, a long history in simulation-heavy labs\u003C\u002Ftd>\u003Ctd>Nanome opens the same coordinates and puts them in a room that several people can walk into, from a browser or a headset. \u003Ca href=\"https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats\">More on supported formats\u003C\u002Fa>.\u003C\u002Ftd>\u003C\u002Ftr>\n  \u003C\u002Ftbody>\n\u003C\u002Ftable>\n\nPyMOL, ChimeraX, Mol* and [VMD](https:\u002F\u002Fwww.ks.uiuc.edu\u002FResearch\u002Fvmd\u002F) are single-user tools driven by a GUI or a script. They're very good at that, and plenty of structural work never needs more than one of them on one screen.\n\n## What Nanome adds\n\n![Two colleagues wearing ultra-thin VR headsets examine the same solid protein surface model with a small ligand visible in its binding pocket floating between them](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_to_view_pdb_files_in_3d_and_the_tools_that_do_it_well_image_3_v4s_aee85e1bd8.png)\n\n\n\nNanome is a collaborative molecular visualization and drug discovery platform that runs on the web and in XR headsets. It opens PDB and SDF structures and fetches from the RCSB PDB, PubChem and DrugBank, so a structure lands in the session without a download and a drag.\n\nTwo things separate it from a desktop viewer. The first is that several people hold one structure at the same time, reaching into the same model from wherever they are instead of watching someone's screen share. That side of it gets a full treatment in [collaborative drug discovery software for remote teams](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fcollaborative-drug-discovery-software-for-remote-teams).\n\nThe second is MARA, an AI copilot inside Nanome that runs scientific tools from plain-English requests. It covers docking, electrostatics with APBS, ADMET prediction, and structure prediction on engines including [AlphaFold 3](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41586-024-07487-w) and [Boltz-2](https:\u002F\u002Fgithub.com\u002Fjwohlwend\u002Fboltz), part of a library of [300+ integrated scientific tools](https:\u002F\u002Fnanome.ai\u002Fintegrations). Every run names the tool it called, the inputs it took and what came back, so the work stays checkable.\n\nNanome also sits alongside the rest of a stack rather than replacing it. [Schrödinger LiveDesign](https:\u002F\u002Fnanome.ai\u002Fintegrations), [Cresset Flare](https:\u002F\u002Fwww.cresset-group.com\u002Fsoftware\u002Fflare\u002F) and [CDD Vault](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fcollaborative-drug-discovery-and-nanome-partnership-announcement) all connect, and [Maestro](https:\u002F\u002Fwww.schrodinger.com\u002Fplatform\u002Fproducts\u002Fmaestro\u002F) `.mae` and `.maegz` files come in as they are, which is how LiveDesign passes a structure across. [More on supported formats](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n\nTurning any of this into something a room can follow is its own question, taken up in [what a modern molecular presentation looks like](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fwhat-a-modern-molecular-presentation-looks-like).\n\nFor a scripted, publication-ready still image on one machine, PyMOL or ChimeraX is the better pick, and Nanome sits well next to either for the shared and immersive parts. The [case studies](https:\u002F\u002Fnanome.ai\u002Fcase-studies) show how groups have split the two on live programs.\n\n## FAQ\n\n**What's the best software for viewing PDB files?**\nFor a look with nothing installed, Mol* in the browser is the quickest route. For scripting and rendered figures, PyMOL or ChimeraX. For a structure several people share across a browser and a headset, with RCSB fetch built in, Nanome.\n\n**How do I view a PDB structure in 3D without installing anything?**\n[molstar.org\u002Fviewer](https:\u002F\u002Fmolstar.org\u002Fviewer) opens a dragged-in `.pdb` file in the browser. Nanome's web app does the same in a browser tab, with no headset involved.\n\n**Can I load a protein and its ligand together?**\nYes. A PDB file usually carries the protein or the complex, an SDF usually carries the small molecule, and Nanome, PyMOL and ChimeraX all read both, so the two sit in one workspace.\n\n**Which formats can Nanome open besides PDB?**\nStructures import as PDB (`.pdb`, `.ent`), mmCIF (`.cif`, `.mmcif`, `.bcif`), SDF (`.sdf`, `.mol`), MOL2, SMILES, XYZ, PQR and PDBQT. Vendor and session files come in too: PyMOL `.pse`, Maestro `.mae` and `.maegz`, and MOE `.moe`. Writing back out is narrower: PDB, SDF or SMILES, one frame at a time, which leaves mmCIF, MAE, MOE and PSE import-only. Simulation frame files and `.dx` maps follow their own rules, set out on the [supported formats page](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n\n**Do I need a headset to use Nanome?**\nNo. The web app runs in a normal browser, and there's a Windows desktop build. Headsets are there for when walking around a structure at full scale is the point: Meta Quest, Apple Vision Pro, HTC Vive Focus 3 and Pico Neo.\n","2026-07-15T01:23:52.242Z","2026-09-01T01:00:00.079Z","2026-09-01T00:33:38.992Z","The best software for viewing PDB files in 3D: PyMOL, ChimeraX, Mol*, and Nanome, which opens PDB and SDF in a browser or XR.","view PDB files, best software for viewing PDB files, how to view a PDB structure in 3D, PDB viewer, PyMOL, ChimeraX, Mol*, Nanome, RCSB PDB","how-to-view-pdb-files-in-3d-and-the-tools-that-do-it-well",{"id":400,"attributes":401},86,{"title":402,"content":403,"createdAt":404,"updatedAt":405,"publishedAt":406,"date":340,"description":407,"keywords":408,"slug":409,"category":17},"How to visualize proteins in VR","To visualize proteins in VR, use Nanome, a collaborative molecular visualization and drug discovery platform that [runs on XR headsets, Windows desktop, and a browser web app](https:\u002F\u002Fnanome.ai\u002Fsetup). You put on a supported headset, open Nanome, load a structure from the RCSB Protein Data Bank (or your own PDB or SDF file), and the protein appears as a full-scale 3D model you can walk around, grab, and rotate with your hands. If you don't own a headset, the same structures open in the Nanome web app.\n\nThe short version: pick a headset, get Nanome, load a structure, choose how it's drawn, then collaborate with others or hand work to [MARA, Nanome's AI copilot](https:\u002F\u002Fnanome.ai\u002Fmara).\n\n## Step by step\n\n![A researcher wearing a slim VR headset holds a ribbon-cartoon protein structure close to their chest, both hands engaged with the model in an open, uncluttered space.](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_to_visualize_proteins_in_vr_image_1_r3_0caa3f79a2.png)\n\n\n\n1. **Pick a headset.** Nanome runs on Meta Quest, Apple Vision Pro, HTC Vive Focus 3, and Pico Neo. There's also a Windows desktop build and a browser web app, and neither of those needs a headset.\n2. **Get Nanome.** Install it on the headset, or open the web app in a browser. Sign in, then create a workspace or join one a colleague set up.\n3. **Load a structure.** Pull a protein from the RCSB PDB by its 4-character code, or import a file of your own. Nanome also reaches PubChem, DrugBank, ChEMBL, UniProt, and the AlphaFold Protein Structure Database.\n4. **Choose representations.** Switch the protein between cartoon ribbons, ball-and-stick, surface, and wireframe. Color by chain, by element, or by whichever property the question turns on.\n5. **Collaborate, or hand work to MARA.** Invite colleagues into the workspace and everyone watches the same molecule at the same moment. Or ask MARA in plain English to run [docking, electrostatics, structure prediction, and more](https:\u002F\u002Fnanome.ai\u002Fagents). With the headset on you can say the request out loud rather than type it, which keeps both hands on the molecule. Once a ligand is seated in the pocket, [how to analyze protein-ligand interactions](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fhow-to-analyze-protein-ligand-interactions) walks through the measurements that come next.\n\n## Supported devices\n\n\u003Ctable class=\"table\">\n  \u003Cthead>\n    \u003Ctr>\u003Cth>Device\u003C\u002Fth>\u003Cth>Type\u003C\u002Fth>\u003Cth>Headset needed\u003C\u002Fth>\u003C\u002Ftr>\n  \u003C\u002Fthead>\n  \u003Ctbody>\n    \u003Ctr>\u003Ctd>Meta Quest\u003C\u002Ftd>\u003Ctd>Standalone VR\u003C\u002Ftd>\u003Ctd>Yes\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Apple Vision Pro\u003C\u002Ftd>\u003Ctd>Mixed reality\u003C\u002Ftd>\u003Ctd>Yes\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>HTC Vive Focus 3\u003C\u002Ftd>\u003Ctd>Standalone VR\u003C\u002Ftd>\u003Ctd>Yes\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Pico Neo\u003C\u002Ftd>\u003Ctd>Standalone VR\u003C\u002Ftd>\u003Ctd>Yes\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Windows desktop\u003C\u002Ftd>\u003Ctd>Desktop app\u003C\u002Ftd>\u003Ctd>No\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Web app (browser)\u003C\u002Ftd>\u003Ctd>Browser\u003C\u002Ftd>\u003Ctd>No\u003C\u002Ftd>\u003C\u002Ftr>\n  \u003C\u002Ftbody>\n\u003C\u002Ftable>\n\nThe web app matters here. Not everyone on a team owns a headset, and a browser link lets a chemist in VR share a live view with a colleague at a laptop. Same structure, same session. That arrangement is the whole subject of our guide to [collaborative drug discovery software for remote teams](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fcollaborative-drug-discovery-software-for-remote-teams).\n\n## Why VR for proteins\n\n![A researcher wearing an ultra-thin VR headset holds a solid protein surface model close to their chest, the binding pocket clearly visible on the floating 3D structure.](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_to_visualize_proteins_in_vr_image_2_r3_944aa42ab0.png)\n\n\n\nA protein is a 3D object, and a flat screen flattens it. In VR you see depth, scale, and the shape of a binding pocket the way your hands understand it. You put a hand into the pocket and turn the ligand yourself, rather than dragging a mouse across 2 axes.\n\nNanome grew out of that idea, and chemistry designed inside it has left the headset and been made. Researchers at [Oak Ridge National Laboratory](https:\u002F\u002Fnanome.ai\u002Fcase-studies\u002Fornl) built a new inhibitor for the SARS-CoV-2 main protease in Nanome, adding a chlorine atom that bound the protease better, and it showed superior inhibition in vitro. The Journal of Medicinal Chemistry published the result. \"This novel chemical structure is different from what has been previously studied by the global community,\" said Dr. Daniel Kneller, the paper's first author. If it clears further development, it will be the first drug ever discovered in virtual reality.\n\nThe rendering holds up under measurement too. In [published benchmarks](https:\u002F\u002Fnanome.ai\u002Fpublications), Nanome's surface rendering beats PyMOL, Chimera, Discovery Studio, and ChimeraX by several fold, and load times stay comparable to 2D tools while both eye views draw at 90+ frames per second.\n\n## Where other tools fit\n\n\u003Cdiv class=\"logo-row\" style=\"display:flex;flex-wrap:wrap;align-items:center;justify-content:center;gap:1.75rem;background:#fff;border-radius:12px;padding:1.25rem 1.5rem;margin:2rem 0\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_to_visualize_proteins_in_vr_image_4_ce3bd63c4b.png\" alt='PyMOL' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_to_visualize_proteins_in_vr_image_6_750a7dc22b.png\" alt='VMD' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_to_visualize_proteins_in_vr_image_8_8671fed7c0.png\" alt='Cresset Flare' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003Cimg src=\"https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_to_visualize_proteins_in_vr_image_9_d3fce4d7df.png\" alt='CDD Vault' style=\"width:120px;height:44px;object-fit:contain;display:inline-block\">\u003C\u002Fdiv>\n\n\n\n![Two colleagues at a plain table study a space-filling protein model on a large monitor, one pointing at the structure with a pen.](https:\u002F\u002Fnanome-cms.s3.us-west-1.amazonaws.com\u002Fhow_to_visualize_proteins_in_vr_image_2_8f65f8febe.png)\n\n\n\n[PyMOL](https:\u002F\u002Fpymol.org), [UCSF ChimeraX](https:\u002F\u002Fwww.rbvi.ucsf.edu\u002Fchimerax\u002F), and [VMD](https:\u002F\u002Fwww.ks.uiuc.edu\u002FResearch\u002Fvmd\u002F) are excellent desktop viewers. They're mostly single-user, driven by a GUI or a script, and drawn on a flat screen. For a precise scripted figure at a desk, they're the right tools for the job.\n\nNanome reads what they write. PDB and mmCIF carry the macromolecules; SDF, MOL2, SMILES, and XYZ carry the small ones; PQR and PDBQT load as well, with PDBQT arriving as PDB and its charges dropped. PyMOL `.pse` sessions open for viewing, with a caveat: the validated PyMOL versions aren't documented, and QM\u002FMM link atoms can break the load. [More on supported formats](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n\nNanome sits alongside the big suites rather than standing in for them. It connects to [Schrödinger LiveDesign](https:\u002F\u002Fnanome.ai\u002Fintegrations), Cresset Flare, and [CDD Vault](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fcollaborative-drug-discovery-and-nanome-partnership-announcement), and it opens their native files: [Maestro](https:\u002F\u002Fwww.schrodinger.com\u002Fplatform\u002Fproducts\u002Fmaestro\u002F) `.mae` and `.maegz`, the same route LiveDesign uses to hand structures over, and [MOE](https:\u002F\u002Fwww.chemcomp.com) `.moe`. Those vendor formats come in for viewing only. On the way back out, Nanome writes PDB, SDF, or SMILES, one frame at a time.\n\n\u003Ctable class=\"table\">\n  \u003Cthead>\n    \u003Ctr>\u003Cth>Tool\u003C\u002Fth>\u003Cth>Strength\u003C\u002Fth>\u003Cth>What Nanome adds\u003C\u002Fth>\u003C\u002Ftr>\n  \u003C\u002Fthead>\n  \u003Ctbody>\n    \u003Ctr>\u003Ctd>Schrödinger Maestro, MOE\u003C\u002Ftd>\u003Ctd>Full desktop comp-chem suites\u003C\u002Ftd>\u003Ctd>Reads their files; MARA runs tools in the same session\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>PyMOL \u002F ChimeraX \u002F VMD\u003C\u002Ftd>\u003Ctd>Scripted desktop rendering, publication figures\u003C\u002Ftd>\u003Ctd>Immersive 3D and a shared live session on top\u003C\u002Ftd>\u003C\u002Ftr>\n    \u003Ctr>\u003Ctd>Browser molecule viewers\u003C\u002Ftd>\u003Ctd>Fast preview of a single structure\u003C\u002Ftd>\u003Ctd>A full workspace, several people at once, and an AI copilot\u003C\u002Ftd>\u003C\u002Ftr>\n  \u003C\u002Ftbody>\n\u003C\u002Ftable>\n\nMore projects like the Oak Ridge one are written up at [nanome.ai\u002Fcase-studies](https:\u002F\u002Fnanome.ai\u002Fcase-studies), and our guide to [what a modern molecular presentation looks like](https:\u002F\u002Fnanome.ai\u002Fblog\u002Fwhat-a-modern-molecular-presentation-looks-like) takes it from there.\n\n## FAQ\n\n**How do I view protein structures in virtual reality?**\nInstall Nanome on a supported headset (Meta Quest, Apple Vision Pro, HTC Vive Focus 3, or Pico Neo), sign in, open a workspace, and load a structure by its RCSB PDB code or from a file of your own. The protein renders at full scale, and you grab it and turn it with your hands.\n\n**Can I visualize proteins without a VR headset?**\nYes. Nanome runs as a Windows desktop app and in the browser. The structures are the same either way, and a laptop user can drop into the live session a headset user is already in.\n\n**What file formats and databases does Nanome support?**\nStructures import as PDB and `.ent`, mmCIF, SDF, MOL and MOL2, SMILES, XYZ, PQR, and PDBQT (which arrives as PDB, charges dropped). Vendor and session files open for viewing: Maestro `.mae` and `.maegz`, MOE `.moe`, PyMOL `.pse`. DX electrostatic maps overlay onto a model that is already loaded. Export is PDB, SDF, or SMILES, single frame, and mmCIF, MAE, MOE, and PSE are import-only. Structures come from the RCSB Protein Data Bank, PubChem, DrugBank, ChEMBL, UniProt, and the AlphaFold Protein Structure Database. The full tiered list lives in [the format docs](https:\u002F\u002Fdocs.nanome.ai\u002Fnanome_web\u002Ffileformats).\n\n**Can I run calculations on a protein while I'm wearing the headset?**\nYes, through MARA, Nanome's AI copilot. Ask in plain English and it will dock a ligand with Smina or DiffDock-L, run electrostatics through APBS, predict ADMET and toxicity, or fold a sequence with Boltz-2, AlphaFold 3, or OpenFold3, drawing on a library of 300+ tools across 26 categories. Each run reports the tool it called, the inputs it took, and what came back, so a colleague can retrace it.\n\n**Does everyone in a session need the same headset?**\nNo. One workspace holds a mix of headset, desktop, and browser users, so a structural biologist in a Quest and a project lead on a laptop can study the same pocket together.\n","2026-07-15T01:23:50.501Z","2026-09-01T01:00:01.081Z","2026-09-01T00:33:40.710Z","How to visualize proteins in VR with Nanome: pick a headset, load a PDB or SDF structure, choose representations, then collaborate or run MARA tools.","how to visualize proteins in VR, how to view protein structures in virtual reality, protein visualization VR, Nanome VR protein viewer, PDB in virtual reality","how-to-visualize-proteins-in-vr",{"pagination":411},{"page":412,"pageSize":413,"pageCount":412,"total":414},1,200,36,1790275902455]