{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,14]],"date-time":"2026-05-14T11:12:19Z","timestamp":1778757139401,"version":"3.51.4"},"reference-count":30,"publisher":"Oxford University Press (OUP)","issue":"Supplement_1","license":[{"start":{"date-parts":[[2020,7,13]],"date-time":"2020-07-13T00:00:00Z","timestamp":1594598400000},"content-version":"vor","delay-in-days":12,"URL":"http:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"}],"funder":[{"DOI":"10.13039\/100000002","name":"National Institutes of Health","doi-asserted-by":"publisher","award":["R01-GM078221"],"award-info":[{"award-number":["R01-GM078221"]}],"id":[{"id":"10.13039\/100000002","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000002","name":"National Institutes of Health","doi-asserted-by":"publisher","award":["T32-GM008403"],"award-info":[{"award-number":["T32-GM008403"]}],"id":[{"id":"10.13039\/100000002","id-type":"DOI","asserted-by":"publisher"}]},{"name":"National Science Foundation Research Experience for Undergraduates","award":["DBI-1659649"],"award-info":[{"award-number":["DBI-1659649"]}]},{"name":"Maryland Advanced Research Computing Cluster"},{"DOI":"10.13039\/100017241","name":"MARCC","doi-asserted-by":"crossref","id":[{"id":"10.13039\/100017241","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2020,7,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Motivation<\/jats:title>\n                    <jats:p>Antibody structure is largely conserved, except for a complementarity-determining region featuring six variable loops. Five of these loops adopt canonical folds which can typically be predicted with existing methods, while the remaining loop (CDR H3) remains a challenge due to its highly diverse set of observed conformations. In recent years, deep neural networks have proven to be effective at capturing the complex patterns of protein structure. This work proposes DeepH3, a deep residual neural network that learns to predict inter-residue distances and orientations from antibody heavy and light chain sequence. The output of DeepH3 is a set of probability distributions over distances and orientation angles between pairs of residues. These distributions are converted to geometric potentials and used to discriminate between decoy structures produced by RosettaAntibody and predict new CDR H3 loop structures de novo.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>When evaluated on the Rosetta antibody benchmark dataset of 49 targets, DeepH3-predicted potentials identified better, same and worse structures [measured by root-mean-squared distance (RMSD) from the experimental CDR H3 loop structure] than the standard Rosetta energy function for 33, 6 and 10 targets, respectively, and improved the average RMSD of predictions by 32.1% (1.4\u2009\u00c5). Analysis of individual geometric potentials revealed that inter-residue orientations were more effective than inter-residue distances for discriminating near-native CDR H3 loops. When applied to de novo prediction of CDR H3 loop structures, DeepH3 achieves an average RMSD of 2.2\u2009\u00b1\u20091.1\u2009\u00c5 on the Rosetta antibody benchmark.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Availability and Implementation<\/jats:title>\n                    <jats:p>DeepH3 source code and pre-trained model parameters are freely available at https:\/\/github.com\/Graylab\/deepH3-distances-orientations.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Supplementary information<\/jats:title>\n                    <jats:p>Supplementary data are available at Bioinformatics online.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1093\/bioinformatics\/btaa457","type":"journal-article","created":{"date-parts":[[2020,5,4]],"date-time":"2020-05-04T15:13:41Z","timestamp":1588605221000},"page":"i268-i275","source":"Crossref","is-referenced-by-count":74,"title":["Geometric potentials from deep learning improve prediction of CDR H3 loop structures"],"prefix":"10.1093","volume":"36","author":[{"given":"Jeffrey A","family":"Ruffolo","sequence":"first","affiliation":[{"name":"Program in Molecular Biophysics, The Johns Hopkins University , Baltimore, MD 21218, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Carlos","family":"Guerra","sequence":"additional","affiliation":[{"name":"Department of Computer Science, George Mason University , Fairfax, VA 22030, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Sai Pooja","family":"Mahajan","sequence":"additional","affiliation":[{"name":"Department of Chemical and Biomolecular Engineering, The Johns Hopkins University , Baltimore, MD 21218, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jeremias","family":"Sulam","sequence":"additional","affiliation":[{"name":"Department of Biomedical Engineering, The Johns Hopkins University , Baltimore, MD 21218, USA"},{"name":"Mathematical Institute for Data Science, The Johns Hopkins University , Baltimore, MD 21218, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jeffrey J","family":"Gray","sequence":"additional","affiliation":[{"name":"Program in Molecular Biophysics, The Johns Hopkins University , Baltimore, MD 21218, USA"},{"name":"Department of Chemical and Biomolecular Engineering, The Johns Hopkins University , Baltimore, MD 21218, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"286","published-online":{"date-parts":[[2020,7,13]]},"reference":[{"key":"2024021913331022000_btaa457-B1","doi-asserted-by":"crossref","first-page":"D432","DOI":"10.1093\/nar\/gku1106","article-title":"PyIgClassify: a database of antibody CDR structural classifications","volume":"43","author":"Adolf-Bryfogle","year":"2014","journal-title":"Nucleic Acids Res"},{"key":"2024021913331022000_btaa457-B2","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":"Alford","year":"2017","journal-title":"J. Chem. Theory Comput"},{"key":"2024021913331022000_btaa457-B3","doi-asserted-by":"crossref","first-page":"1553","DOI":"10.1002\/prot.24567","article-title":"Second antibody modeling assessment (AMA-II","volume":"82","author":"Almagro","year":"2014","journal-title":"Prot. Struct. Funct. Bioinform"},{"key":"2024021913331022000_btaa457-B4","doi-asserted-by":"crossref","first-page":"1636","DOI":"10.1002\/prot.24595","article-title":"Automated Aufbau of antibody structures from given sequences using Macromoltek\u2019s SmrtMolAntibody","volume":"82","author":"Berrondo","year":"2014","journal-title":"Prot. Struct. Funct. Bioinform"},{"key":"2024021913331022000_btaa457-B5","doi-asserted-by":"crossref","first-page":"877","DOI":"10.1038\/342877a0","article-title":"Conformations of immunoglobulin hypervariable regions","volume":"342","author":"Chothia","year":"1989","journal-title":"Nature"},{"key":"2024021913331022000_btaa457-B6","doi-asserted-by":"crossref","first-page":"47","DOI":"10.1002\/pro.2389","article-title":"Relaxation of backbone bond geometry improves protein energy landscape modeling","volume":"23","author":"Conway","year":"2014","journal-title":"Prot. Sci"},{"key":"2024021913331022000_btaa457-B7","doi-asserted-by":"crossref","first-page":"D1140","DOI":"10.1093\/nar\/gkt1043","article-title":"SAbDab: the structural antibody database","volume":"42","author":"Dunbar","year":"2014","journal-title":"Nucleic Acids Res"},{"key":"2024021913331022000_btaa457-B8","doi-asserted-by":"crossref","first-page":"1422","DOI":"10.1016\/j.celrep.2019.03.097","article-title":"Comprehensive profiling of HIV antibody evolution","volume":"27","author":"Eshleman","year":"2019","journal-title":"Cell Rep"},{"key":"2024021913331022000_btaa457-B9","doi-asserted-by":"crossref","first-page":"1583","DOI":"10.1002\/prot.24604","article-title":"Automated antibody structure prediction using Accelrys tools: results and best practices","volume":"82","author":"Fasnacht","year":"2014","journal-title":"Prot. Struct. Funct. Bioinform"},{"key":"2024021913331022000_btaa457-B66210065","first-page":"770","author":"He","year":"2016"},{"key":"2024021913331022000_btaa457-B10","doi-asserted-by":"crossref","first-page":"2801","DOI":"10.1093\/bioinformatics\/bty1037","article-title":"Smooth orientation-dependent scoring function for coarse-grained protein quality assessment","volume":"35","author":"Karasikov","year":"2019","journal-title":"Bioinformatics"},{"key":"2024021913331022000_btaa457-B11","doi-asserted-by":"crossref","first-page":"3013","DOI":"10.1093\/bioinformatics\/btz026","article-title":"KORP: knowledge-based 6D potential for fast protein and loop modeling","volume":"35","author":"L\u00f3pez-Blanco","year":"2019","journal-title":"Bioinformatics"},{"key":"2024021913331022000_btaa457-B12","doi-asserted-by":"crossref","first-page":"1599","DOI":"10.1002\/prot.24576","article-title":"Assessment of fully automated antibody homology modeling protocols in molecular operating environment","volume":"82","author":"Maier","year":"2014","journal-title":"Prot. Struct. Funct. Bioinform"},{"key":"2024021913331022000_btaa457-B13","doi-asserted-by":"crossref","first-page":"551","DOI":"10.1038\/nmeth0809-551","article-title":"Sub-angstrom accuracy in protein loop reconstruction by robotics-inspired conformational sampling","volume":"6","author":"Mandell","year":"2009","journal-title":"Nat. Methods"},{"key":"2024021913331022000_btaa457-B14","doi-asserted-by":"crossref","first-page":"409","DOI":"10.1093\/protein\/gzw013","article-title":"Improved prediction of antibody VL\u2013VH orientation","volume":"29","author":"Marze","year":"2016","journal-title":"Prot. Eng. Des. Selection"},{"key":"2024021913331022000_btaa457-B15","doi-asserted-by":"crossref","first-page":"228","DOI":"10.1016\/j.jmb.2010.10.030","article-title":"A new clustering of antibody CDR loop conformations","volume":"406","author":"North","year":"2011","journal-title":"J. Mol. Biol"},{"key":"2024021913331022000_btaa457-B16","first-page":"8024","author":"Paszke","year":"2019"},{"key":"2024021913331022000_btaa457-B17","doi-asserted-by":"crossref","first-page":"4890","DOI":"10.4049\/jimmunol.1201493","article-title":"A systematic comparison of free and bound antibodies reveals binding-related conformational changes","volume":"189","author":"Sela-Culang","year":"2012","journal-title":"J. Immunol"},{"key":"2024021913331022000_btaa457-B18","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":"Senior","year":"2020","journal-title":"Nature"},{"key":"2024021913331022000_btaa457-B19","doi-asserted-by":"crossref","first-page":"1624","DOI":"10.1002\/prot.24591","article-title":"High-resolution modeling of antibody structures by a combination of bioinformatics, expert knowledge, and molecular simulations","volume":"82","author":"Shirai","year":"2014","journal-title":"Prot. Struct. Funct. Bioinform"},{"key":"2024021913331022000_btaa457-B20","doi-asserted-by":"crossref","first-page":"e63090","DOI":"10.1371\/journal.pone.0063090","article-title":"Improvements to robotics-inspired conformational sampling in Rosetta","volume":"8","author":"Stein","year":"2013","journal-title":"PLoS One"},{"key":"2024021913331022000_btaa457-B21","doi-asserted-by":"crossref","first-page":"e1005324","DOI":"10.1371\/journal.pcbi.1005324","article-title":"Accurate de novo prediction of protein contact map by ultra-deep learning model","volume":"13","author":"Wang","year":"2017","journal-title":"PLoS Comput. Biol"},{"key":"2024021913331022000_btaa457-B22","doi-asserted-by":"crossref","first-page":"67","DOI":"10.1002\/prot.25377","article-title":"Analysis of deep learning methods for blind protein contact prediction in CASP12","volume":"86","author":"Wang","year":"2018","journal-title":"Prot. Struct. Funct. Bioinform"},{"key":"2024021913331022000_btaa457-B23","doi-asserted-by":"crossref","first-page":"505","DOI":"10.4049\/jimmunol.1601137","article-title":"Accurate structure prediction of CDR H3 loops enabled by a novel structure-based C-terminal constraint","volume":"198","author":"Weitzner","year":"2017","journal-title":"J. Immunol"},{"key":"2024021913331022000_btaa457-B24","doi-asserted-by":"crossref","first-page":"1611","DOI":"10.1002\/prot.24534","article-title":"Blind prediction performance of RosettaAntibody 3.0: grafting, relaxation, kinematic loop modeling, and full CDR optimization","volume":"82","author":"Weitzner","year":"2014","journal-title":"Prot. Struct. Funct. Bioinform"},{"key":"2024021913331022000_btaa457-B25","doi-asserted-by":"crossref","first-page":"401","DOI":"10.1038\/nprot.2016.180","article-title":"Modeling and docking of antibody structures with Rosetta","volume":"12","author":"Weitzner","year":"2017","journal-title":"Nat. Protocols"},{"key":"2024021913331022000_btaa457-B26","doi-asserted-by":"crossref","first-page":"1593","DOI":"10.1126\/science.1207532","article-title":"Focused evolution of HIV-1 neutralizing antibodies revealed by structures and deep sequencing","volume":"333","author":"Wu","year":"2011","journal-title":"Science"},{"key":"2024021913331022000_btaa457-B27","doi-asserted-by":"crossref","first-page":"16856","DOI":"10.1073\/pnas.1821309116","article-title":"Distance-based protein folding powered by deep learning","volume":"116","author":"Xu","year":"2019","journal-title":"Proc. Natl. Acad. Sci"},{"key":"2024021913331022000_btaa457-B28","first-page":"1496","author":"Yang","year":"2020"},{"key":"2024021913331022000_btaa457-B29","doi-asserted-by":"crossref","first-page":"1646","DOI":"10.1002\/prot.24551","article-title":"Antibody structure determination using a combination of homology modeling, energy-based refinement, and loop prediction","volume":"82","author":"Zhu","year":"2014","journal-title":"Prot. Struct. Funct. Bioinform"}],"container-title":["Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/36\/Supplement_1\/i268\/56702806\/bioinformatics_36_supplement1_i268.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/36\/Supplement_1\/i268\/56702806\/bioinformatics_36_supplement1_i268.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,2,19]],"date-time":"2024-02-19T08:41:14Z","timestamp":1708332074000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article\/36\/Supplement_1\/i268\/5870520"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2020,7,1]]},"references-count":30,"journal-issue":{"issue":"Supplement_1","published-print":{"date-parts":[[2020,7,1]]}},"URL":"https:\/\/doi.org\/10.1093\/bioinformatics\/btaa457","relation":{"has-preprint":[{"id-type":"doi","id":"10.1101\/2020.02.09.940254","asserted-by":"object"}]},"ISSN":["1367-4803","1367-4811"],"issn-type":[{"value":"1367-4803","type":"print"},{"value":"1367-4811","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2020,7]]},"published":{"date-parts":[[2020,7,1]]}}}