{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,19]],"date-time":"2026-04-19T04:15:27Z","timestamp":1776572127563,"version":"3.51.2"},"update-to":[{"DOI":"10.1371\/journal.pcbi.1010271","type":"new_version","label":"New version","source":"publisher","updated":{"date-parts":[[2022,7,8]],"date-time":"2022-07-08T00:00:00Z","timestamp":1657238400000}}],"reference-count":56,"publisher":"Public Library of Science (PLoS)","issue":"6","license":[{"start":{"date-parts":[[2022,6,27]],"date-time":"2022-06-27T00:00:00Z","timestamp":1656288000000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/100006521","name":"School of Medicine, Stanford University","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100006521","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100006192","name":"Advanced Scientific Computing Research","doi-asserted-by":"publisher","award":["SciDAC"],"award-info":[{"award-number":["SciDAC"]}],"id":[{"id":"10.13039\/100006192","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100014498","name":"Chemistry, Engineering and Medicine for Human Health, Stanford University","doi-asserted-by":"publisher","award":["ChEM-H"],"award-info":[{"award-number":["ChEM-H"]}],"id":[{"id":"10.13039\/100014498","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000057","name":"National Institute of General Medical Sciences","doi-asserted-by":"publisher","award":["T32GM120007."],"award-info":[{"award-number":["T32GM120007."]}],"id":[{"id":"10.13039\/100000057","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100006785","name":"Google","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100006785","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["www.ploscompbiol.org"],"crossmark-restriction":false},"short-container-title":["PLoS Comput Biol"],"abstract":"<jats:p>While deep learning models have seen increasing applications in protein science, few have been implemented for protein backbone generation\u2014an important task in structure-based problems such as active site and interface design. We present a new approach to building class-specific backbones, using a variational auto-encoder to directly generate the 3D coordinates of immunoglobulins. Our model is torsion- and distance-aware, learns a high-resolution embedding of the dataset, and generates novel, high-quality structures compatible with existing design tools. We show that the Ig-VAE can be used with Rosetta to create a computational model of a SARS-CoV2-RBD binder via latent space sampling. We further demonstrate that the model\u2019s generative prior is a powerful tool for guiding computational protein design, motivating a new paradigm under which backbone design is solved as constrained optimization problem in the latent space of a generative model.<\/jats:p>","DOI":"10.1371\/journal.pcbi.1010271","type":"journal-article","created":{"date-parts":[[2022,6,27]],"date-time":"2022-06-27T13:36:45Z","timestamp":1656337005000},"page":"e1010271","update-policy":"https:\/\/doi.org\/10.1371\/journal.pcbi.corrections_policy","source":"Crossref","is-referenced-by-count":100,"title":["Ig-VAE: Generative modeling of protein structure by direct 3D coordinate generation"],"prefix":"10.1371","volume":"18","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-2704-8464","authenticated-orcid":true,"given":"Raphael R.","family":"Eguchi","sequence":"first","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8871-9682","authenticated-orcid":true,"given":"Christian A.","family":"Choe","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7948-2895","authenticated-orcid":true,"given":"Po-Ssu","family":"Huang","sequence":"additional","affiliation":[]}],"member":"340","published-online":{"date-parts":[[2022,6,27]]},"reference":[{"issue":"7724","key":"pcbi.1010271.ref001","doi-asserted-by":"crossref","first-page":"485","DOI":"10.1038\/s41586-018-0509-0","article-title":"De novo design of a fluorescence-activating \u03b2-barrel","volume":"561","author":"J Dou","year":"2018","journal-title":"Nature"},{"issue":"1","key":"pcbi.1010271.ref002","doi-asserted-by":"crossref","first-page":"29","DOI":"10.1038\/nchembio.1966","article-title":"De novo design of a four-fold symmetricTIM-barrel protein with atomic-level accuracy","volume":"12","author":"P Huang","year":"2015","journal-title":"Nature Chemical Biology"},{"issue":"7738","key":"pcbi.1010271.ref003","doi-asserted-by":"crossref","first-page":"186","DOI":"10.1038\/s41586-018-0830-7","article-title":"De novo design of potent and selective mimics of IL-2 and IL-15","volume":"565","author":"DA Silva","year":"2019","journal-title":"Nature"},{"issue":"6059","key":"pcbi.1010271.ref004","doi-asserted-by":"crossref","first-page":"1097","DOI":"10.1126\/science.1213256","article-title":"A Potent and Broad Neutralizing Antibody Recognizes and Penetrates the HIV Glycan Shield","volume":"334","author":"R Pejchal","year":"2011","journal-title":"Science"},{"issue":"7192","key":"pcbi.1010271.ref005","doi-asserted-by":"crossref","first-page":"190","DOI":"10.1038\/nature06879","article-title":"Kemp elimination catalysts by computational enzyme design","volume":"453","author":"D R\u00f6thlisberger","year":"2008","journal-title":"Nature"},{"issue":"7768","key":"pcbi.1010271.ref006","doi-asserted-by":"crossref","first-page":"205","DOI":"10.1038\/s41586-019-1432-8","article-title":"De novo design of bioactive protein switches","volume":"572","author":"RA Langan","year":"2019","journal-title":"Nature"},{"issue":"13","key":"pcbi.1010271.ref007","doi-asserted-by":"crossref","first-page":"7208","DOI":"10.1073\/pnas.1914808117","article-title":"Computational design of closely related proteins that adopt two well-defined but structurally divergent folds","volume":"117","author":"KY Wei","year":"2020","journal-title":"Proceedings of the National Academy of Sciences"},{"issue":"6379","key":"pcbi.1010271.ref008","doi-asserted-by":"crossref","first-page":"1042","DOI":"10.1126\/science.aaq1739","article-title":"Accurate computational design of multipass transmembrane proteins","volume":"359","author":"P Lu","year":"2018","journal-title":"Science"},{"issue":"7620","key":"pcbi.1010271.ref009","doi-asserted-by":"crossref","first-page":"320","DOI":"10.1038\/nature19946","article-title":"The coming of age of de novo protein design","volume":"537","author":"PS Huang","year":"2016","journal-title":"Nature"},{"key":"pcbi.1010271.ref010","doi-asserted-by":"crossref","first-page":"545","DOI":"10.1016\/B978-0-12-381270-4.00019-6","volume-title":"Computer Methods, Part C. vol. 487 of Methods in Enzymology","author":"A Leaver-Fay","year":"2011"},{"key":"pcbi.1010271.ref011","article-title":"Protein Sequence Design with a Learned Potential","author":"N Anand","year":"2020","journal-title":"Bioinformatics"},{"issue":"6","key":"pcbi.1010271.ref012","doi-asserted-by":"crossref","first-page":"1740","DOI":"10.1093\/bioinformatics\/btz650","article-title":"Multi-scale structural analysis of proteins by deep semantic segmentation","volume":"36","author":"RR Eguchi","year":"2020","journal-title":"Bioinformatics"},{"key":"pcbi.1010271.ref013","article-title":"Deciphering interaction fingerprints from protein molecular surfaces using geometric deep learning","author":"P Gainza","year":"2019","journal-title":"Nature Methods"},{"issue":"7792","key":"pcbi.1010271.ref014","doi-asserted-by":"crossref","first-page":"706","DOI":"10.1038\/s41586-019-1923-7","article-title":"Improved protein structure prediction using potentials from deep learning","volume":"577","author":"AW Senior","year":"2020","journal-title":"Nature"},{"issue":"3","key":"pcbi.1010271.ref015","doi-asserted-by":"crossref","first-page":"1496","DOI":"10.1073\/pnas.1914677117","article-title":"Improved protein structure prediction using predicted interresidue orientations","volume":"117","author":"J Yang","year":"2020","journal-title":"Proceedings of the National Academy of Sciences"},{"issue":"12","key":"pcbi.1010271.ref016","doi-asserted-by":"crossref","first-page":"1315","DOI":"10.1038\/s41592-019-0598-1","article-title":"Unified rational protein engineering with sequence-based deep representation learning","volume":"16","author":"EC Alley","year":"2019","journal-title":"Nature Methods"},{"key":"pcbi.1010271.ref017","article-title":"ProGen: Language Modeling for Protein Generation.","author":"A Madani","year":"2020","journal-title":"Synthetic Biology"},{"issue":"10","key":"pcbi.1010271.ref018","doi-asserted-by":"crossref","first-page":"816","DOI":"10.1038\/s41592-018-0138-4","article-title":"Deep generative models of genetic variation capture the effects of mutations","volume":"15","author":"AJ Riesselman","year":"2018","journal-title":"Nature Methods"},{"key":"pcbi.1010271.ref019","article-title":"Accelerating Protein Design Using Autoregressive Generative Models.","author":"A Riesselman","year":"2019","journal-title":"bioRxiv."},{"key":"pcbi.1010271.ref020","first-page":"7494","volume-title":"Advances in Neural Information Processing Systems 31","author":"N Anand","year":"2018"},{"key":"pcbi.1010271.ref021","volume-title":"Fully differentiable full-atom protein backbone generation","author":"N Anand","year":"2019"},{"key":"pcbi.1010271.ref022","doi-asserted-by":"crossref","first-page":"136","DOI":"10.1016\/j.cbpa.2021.08.004","article-title":"Structure-based protein design with deep learning","volume":"65","author":"S Ovchinnikov","year":"2021","journal-title":"Current Opinion in Chemical Biology"},{"key":"pcbi.1010271.ref023","article-title":"VAE, a Geometric Convolutional VAE for ProteinStructure Generation.","author":"H Huang","year":"2021","journal-title":"arXiv"},{"key":"pcbi.1010271.ref024","article-title":"Generating Tertiary Protein Structures via an Interpretative Variational Autoencoder.","author":"X Guo","year":"2020","journal-title":"arXiv:200407119 cs, q-bio, stat"},{"key":"pcbi.1010271.ref025","article-title":"Auto-Encoding Variational Bayes","author":"DP Kingma","year":"2014","journal-title":"arXiv:13126114 cs, stat"},{"key":"pcbi.1010271.ref026","article-title":"Generating valid Euclidean distance matrices.","author":"M Hoffmann","journal-title":"arXiv:191003131 cs, stat. 2019 Nov. ArXiv"},{"issue":"20","key":"pcbi.1010271.ref027","doi-asserted-by":"crossref","first-page":"10701","DOI":"10.1073\/pnas.170297297","article-title":"Directed evolution of antibody fragments with monovalent femtomolar antigen-binding affinity","volume":"97","author":"ET Boder","year":"2000","journal-title":"Proceedings of the National Academy of Sciences"},{"issue":"11","key":"pcbi.1010271.ref028","doi-asserted-by":"crossref","first-page":"2901","DOI":"10.1110\/ps.051709505","article-title":"Structure of a shark IgNAR antibody variable domain and modeling of an early-developmental isotype","volume":"14","author":"VA Streltsov","year":"2005","journal-title":"Protein Science"},{"key":"pcbi.1010271.ref029","article-title":"Neutralizing nanobodies bind SARS-CoV-2 spike RBD and block interaction with ACE2","author":"J Huo","year":"2020","journal-title":"Nature Structural & Molecular Biology"},{"issue":"8","key":"pcbi.1010271.ref030","doi-asserted-by":"crossref","first-page":"e1001120","DOI":"10.1371\/journal.pbio.1001120","article-title":"A Case Of Convergence: Why Did a Simple Alternative to Canonical Antibodies Arise in Sharks and Camels","volume":"9","author":"MF Flajnik","year":"2011","journal-title":"PLoS Biology"},{"issue":"8","key":"pcbi.1010271.ref031","doi-asserted-by":"crossref","first-page":"e24109","DOI":"10.1371\/journal.pone.0024109","article-title":"RosettaRemodel: A Generalized Framework for Flexible Backbone Protein Design.","volume":"6","author":"PS Huang","year":"2011","journal-title":"PLoS ONEAug"},{"issue":"7423","key":"pcbi.1010271.ref032","doi-asserted-by":"crossref","first-page":"222","DOI":"10.1038\/nature11600","article-title":"Principles for designing ideal protein structures","volume":"491","author":"N Koga","year":"2012","journal-title":"Nature"},{"key":"pcbi.1010271.ref033","article-title":"AbDb: antibody structure database\u2014a database of PDB-derived antibody structures.","author":"S Ferdous","year":"2018","journal-title":"Database"},{"issue":"1","key":"pcbi.1010271.ref034","doi-asserted-by":"crossref","first-page":"9","DOI":"10.2478\/jtim-2020-0003","article-title":"Potential role of ACE2 in coronavirus disease 2019 (COVID-19) prevention and management.","volume":"8","author":"M Liu","year":"2020","journal-title":"Journal of Translational Internal Medicine"},{"issue":"7625","key":"pcbi.1010271.ref035","doi-asserted-by":"crossref","first-page":"329","DOI":"10.1038\/nature19791","article-title":"Accurate de novo design of hyperstable constrained peptides","volume":"538","author":"G Bhardwaj","year":"2016","journal-title":"Nature"},{"issue":"6","key":"pcbi.1010271.ref036","doi-asserted-by":"crossref","first-page":"e20161","DOI":"10.1371\/journal.pone.0020161","article-title":"RosettaScripts: A Scripting Language Interface to the Rosetta Macromolecular Modeling Suite","volume":"6","author":"SJ Fleishman","year":"2011","journal-title":"PLoS ONE."},{"key":"pcbi.1010271.ref037","first-page":"2579","article-title":"Visualizing Data using t-SNE","volume":"9","author":"L van der Maaten","year":"2008","journal-title":"Journal of Machine Learning Research"},{"issue":"11","key":"pcbi.1010271.ref038","doi-asserted-by":"crossref","first-page":"559","DOI":"10.1080\/14786440109462720","article-title":"LIII. On lines and planes of closest fit to systems of points in space.","volume":"2","author":"S KPFR","year":"1901","journal-title":"The London, Edinburgh, and Dublin Philosophical Magazine and Journal of Science"},{"key":"pcbi.1010271.ref039","article-title":"Camelid Single-Domain Antibodies: Historical Perspective and Future Outlook.","author":"M. Arbabi-Ghahroudi","year":"2017","journal-title":"Frontiers in Immunology"},{"key":"pcbi.1010271.ref040","doi-asserted-by":"crossref","first-page":"341","DOI":"10.1006\/rwei.1999.0089","volume-title":"Encyclopedia of Immunology (Second Edition)","author":"JL Preud\u2019homme","year":"1998","edition":"2"},{"key":"pcbi.1010271.ref041","doi-asserted-by":"crossref","first-page":"W363","DOI":"10.1093\/nar\/gki481","article-title":"PatchDock and SymmDock: servers for rigid and symmetric docking","volume":"33","author":"D Schneidman-Duhovny","year":"2005","journal-title":"Nucleic Acids Research"},{"issue":"8","key":"pcbi.1010271.ref042","doi-asserted-by":"crossref","first-page":"e22477","DOI":"10.1371\/journal.pone.0022477","article-title":"Benchmarking and Analysis of Protein Docking Performance in Rosetta v3.2.","volume":"6","author":"S Chaudhury","year":"2011","journal-title":"PLoS ONE"},{"issue":"1","key":"pcbi.1010271.ref043","doi-asserted-by":"crossref","first-page":"281","DOI":"10.1016\/S0022-2836(03)00670-3","article-title":"Protein\u2013Protein Docking with Simultaneous Optimization of Rigid-body Displacement and Side-chain Conformations","volume":"331","author":"JJ Gray","year":"2003","journal-title":"Journal of Molecular Biology"},{"key":"pcbi.1010271.ref044","article-title":"Precise Recovery of Latent Vectors from Generative Adversarial Networks","author":"ZC Lipton","year":"2017","journal-title":"arXiv:170204782 cs, stat"},{"key":"pcbi.1010271.ref045","article-title":"Generalized Latent Variable Recovery for Generative Adversarial Networks.","author":"N Egan","year":"2018","journal-title":"arXiv:181003764 cs, stat."},{"issue":"4","key":"pcbi.1010271.ref046","doi-asserted-by":"crossref","first-page":"946","DOI":"10.1006\/jmbi.1993.1648","article-title":"Shape Complementarity at Protein\/Protein Interfaces","volume":"234","author":"MC Lawrence","year":"1993","journal-title":"Journal of Molecular Biology"},{"issue":"6","key":"pcbi.1010271.ref047","doi-asserted-by":"crossref","first-page":"3031","DOI":"10.1021\/acs.jctc.7b00125","article-title":"The Rosetta All-Atom Energy Function for Macromolecular Modeling and Design","volume":"13","author":"RF Alford","year":"2017","journal-title":"Journal of Chemical Theory and Computation"},{"issue":"3","key":"pcbi.1010271.ref048","doi-asserted-by":"crossref","first-page":"eaax7379","DOI":"10.1126\/sciadv.aax7379","article-title":"Structure-guided discovery of a single-domain antibody agonist against human apelin receptor","volume":"6","author":"Y Ma","year":"2020","journal-title":"Sci Adv"},{"issue":"7","key":"pcbi.1010271.ref049","first-page":"1311","article-title":"The H3 loop of antibodies shows unique structural characteristics. Proteins: Structure, Function, and","volume":"85","author":"C Regep","year":"2017","journal-title":"Bioinformatics"},{"issue":"2","key":"pcbi.1010271.ref050","doi-asserted-by":"crossref","first-page":"100406","DOI":"10.1016\/j.patter.2021.100406","article-title":"Antibody structure prediction using interpretable deep learning.","volume":"3","author":"JA Ruffolo","year":"2022","journal-title":"Patterns"},{"issue":"7889","key":"pcbi.1010271.ref051","doi-asserted-by":"crossref","first-page":"547","DOI":"10.1038\/s41586-021-04184-w","article-title":"De novo protein design by deep network hallucination","volume":"600","author":"I Anishchenko","year":"2021","journal-title":"Nature"},{"key":"pcbi.1010271.ref052","article-title":"Iterative Refinement Graph Neural Network for Antibody Sequence-Structure Co-design","author":"W Jin","year":"2021","journal-title":"arXiv"},{"issue":"11","key":"pcbi.1010271.ref053","doi-asserted-by":"crossref","first-page":"2634","DOI":"10.1002\/bit.23220","article-title":"Evaluation of protein engineering and process optimization approaches to enhance antibody drug manufacturability","volume":"108","author":"GP Conley","year":"2011","journal-title":"Biotechnology and Bioengineering"},{"issue":"6","key":"pcbi.1010271.ref054","doi-asserted-by":"crossref","first-page":"580","DOI":"10.4161\/mabs.1.6.9773","article-title":"Predictive tools for stabilization of therapeutic proteins","volume":"1","author":"V Voynov","year":"2009","journal-title":"mAbs"},{"issue":"2","key":"pcbi.1010271.ref055","doi-asserted-by":"crossref","first-page":"113","DOI":"10.1007\/s12033-008-9049-4","article-title":"Post-translational Modifications of Recombinant Proteins: Significance for Biopharmaceuticals","volume":"39","author":"N Jenkins","year":"2008","journal-title":"Molecular Biotechnology"},{"issue":"27","key":"pcbi.1010271.ref056","doi-asserted-by":"crossref","first-page":"10879","DOI":"10.1073\/pnas.1202866109","article-title":"General strategy for the generation of human antibody variable domains with increased aggregation resistance","volume":"109","author":"K Dudgeon","year":"2012","journal-title":"Proceedings of the National Academy of Sciences"}],"updated-by":[{"DOI":"10.1371\/journal.pcbi.1010271","type":"new_version","label":"New version","source":"publisher","updated":{"date-parts":[[2022,7,8]],"date-time":"2022-07-08T00:00:00Z","timestamp":1657238400000}}],"container-title":["PLOS Computational Biology"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pcbi.1010271","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,7,8]],"date-time":"2022-07-08T14:23:30Z","timestamp":1657290210000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pcbi.1010271"}},"subtitle":[],"editor":[{"given":"Joanna","family":"Slusky","sequence":"first","affiliation":[]}],"short-title":[],"issued":{"date-parts":[[2022,6,27]]},"references-count":56,"journal-issue":{"issue":"6","published-online":{"date-parts":[[2022,6,27]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pcbi.1010271","relation":{"has-preprint":[{"id-type":"doi","id":"10.1101\/2020.08.07.242347","asserted-by":"object"}]},"ISSN":["1553-7358"],"issn-type":[{"value":"1553-7358","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,6,27]]}}}