{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,13]],"date-time":"2026-05-13T00:32:20Z","timestamp":1778632340860,"version":"3.51.4"},"reference-count":46,"publisher":"Oxford University Press (OUP)","issue":"18","license":[{"start":{"date-parts":[[2019,2,13]],"date-time":"2019-02-13T00:00:00Z","timestamp":1550016000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/academic.oup.com\/journals\/pages\/open_access\/funder_policies\/chorus\/standard_publication_model"}],"funder":[{"DOI":"10.13039\/100007579","name":"Auburn University","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100007579","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2019,9,15]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:sec><jats:title>Motivation<\/jats:title><jats:p>Protein structure refinement aims to bring moderately accurate template-based protein models closer to the native state through conformational sampling. However, guiding the sampling towards the native state by effectively using restraints remains a major issue in structure refinement.<\/jats:p><\/jats:sec><jats:sec><jats:title>Results<\/jats:title><jats:p>Here, we develop a machine learning based restrained relaxation protocol that uses deep discriminative learning based binary classifiers to predict multi-resolution probabilistic restraints from the starting structure and subsequently converts these restraints to be integrated into Rosetta all-atom energy function as additional scoring terms during structure refinement. We use four restraint resolutions as adopted in GDT-HA (0.5, 1, 2 and 4\u2009\u00c5), centered on the C\u03b1 atom of each residue that are predicted by ensemble of four deep discriminative classifiers trained using combinations of sequence and structure-derived features as well as several energy terms from Rosetta centroid scoring function. The proposed method, refineD, has been found to produce consistent and substantial structural refinement through the use of cumulative and non-cumulative restraints on 150 benchmarking targets. refineD outperforms unrestrained relaxation strategy or relaxation that is restrained to starting structures using the FastRelax application of Rosetta or atomic-level energy minimization based ModRefiner method as well as molecular dynamics (MD) simulation based FG-MD protocol. Furthermore, by adjusting restraint resolutions, the method addresses the tradeoff that exists between degree and consistency of refinement. These results demonstrate a promising new avenue for improving accuracy of template-based protein models by effectively guiding conformational sampling during structure refinement through the use of machine learning based restraints.<\/jats:p><\/jats:sec><jats:sec><jats:title>Availability and implementation<\/jats:title><jats:p>http:\/\/watson.cse.eng.auburn.edu\/refineD\/.<\/jats:p><\/jats:sec><jats:sec><jats:title>Supplementary information<\/jats:title><jats:p>Supplementary data are available at Bioinformatics online.<\/jats:p><\/jats:sec>","DOI":"10.1093\/bioinformatics\/btz101","type":"journal-article","created":{"date-parts":[[2019,2,13]],"date-time":"2019-02-13T21:17:19Z","timestamp":1550092639000},"page":"3320-3328","source":"Crossref","is-referenced-by-count":32,"title":["refineD: improved protein structure refinement using machine learning based restrained relaxation"],"prefix":"10.1093","volume":"35","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-9630-0141","authenticated-orcid":false,"given":"Debswapna","family":"Bhattacharya","sequence":"first","affiliation":[{"name":"Department of Computer Science and Software Engineering, Auburn University , Auburn, AL, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"286","published-online":{"date-parts":[[2019,2,13]]},"reference":[{"key":"2023013108053317100_btz101-B1","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":"2023013108053317100_btz101-B2","doi-asserted-by":"crossref","first-page":"3389","DOI":"10.1093\/nar\/25.17.3389","article-title":"Gapped BLAST and PSI-BLAST: a new generation of protein database search programs","volume":"25","author":"Altschul","year":"1997","journal-title":"Nucleic Acids Res"},{"key":"2023013108053317100_btz101-B3","doi-asserted-by":"crossref","first-page":"119","DOI":"10.1002\/prot.24167","article-title":"3Drefine: consistent protein structure refinement by optimizing hydrogen bonding network and atomic-level energy minimization","volume":"81","author":"Bhattacharya","year":"2013","journal-title":"Proteins Struct. Funct. Bioinf"},{"key":"2023013108053317100_btz101-B4","doi-asserted-by":"crossref","first-page":"e69648.","DOI":"10.1371\/journal.pone.0069648","article-title":"i3Drefine software for protein 3D structure refinement and its assessment in CASP10","volume":"8","author":"Bhattacharya","year":"2013","journal-title":"PLoS One"},{"key":"2023013108053317100_btz101-B5","first-page":"106","author":"Bhattacharya","year":"2013"},{"key":"2023013108053317100_btz101-B6","doi-asserted-by":"crossref","first-page":"W406","DOI":"10.1093\/nar\/gkw336","article-title":"3Drefine: an interactive web server for efficient protein structure refinement","volume":"44","author":"Bhattacharya","year":"2016","journal-title":"Nucleic Acids Res"},{"key":"2023013108053317100_btz101-B7","first-page":"484","article-title":"Refinement of comparative-modeling structures by multicanonical molecular dynamics","volume":"14","author":"Cao","year":"2003","journal-title":"Genome Inf"},{"key":"2023013108053317100_btz101-B8","doi-asserted-by":"crossref","first-page":"676","DOI":"10.1016\/j.drudis.2009.04.006","article-title":"Homology modeling in drug discovery: current trends and applications","volume":"14","author":"Cavasotto","year":"2009","journal-title":"Drug Disc. Today"},{"key":"2023013108053317100_btz101-B9","doi-asserted-by":"crossref","first-page":"689","DOI":"10.1093\/bioinformatics\/btq007","article-title":"PyRosetta: a script-based interface for implementing molecular modeling algorithms using Rosetta","volume":"26","author":"Chaudhury","year":"2010","journal-title":"Bioinformatics"},{"key":"2023013108053317100_btz101-B10","doi-asserted-by":"crossref","first-page":"922","DOI":"10.1002\/prot.21345","article-title":"Can molecular dynamics simulations provide high-resolution refinement of protein structure?","volume":"67","author":"Chen","year":"2007","journal-title":"Proteins Struct. Funct. Bioinf"},{"key":"2023013108053317100_btz101-B11","doi-asserted-by":"crossref","first-page":"302","DOI":"10.1002\/prot.24939","article-title":"Protein structure refinement with adaptively restrained homologous replicas","volume":"84","author":"Della Corte","year":"2016","journal-title":"Proteins Struct. Funct. Bioinf"},{"key":"2023013108053317100_btz101-B12","doi-asserted-by":"crossref","first-page":"378","DOI":"10.1093\/bioinformatics\/btv601","article-title":"3DRobot: automated generation of diverse and well-packed protein structure decoys","volume":"32","author":"Deng","year":"2016","journal-title":"Bioinformatics"},{"key":"2023013108053317100_btz101-B13","doi-asserted-by":"crossref","DOI":"10.1002\/wcms.1307","article-title":"Computational protein structure refinement: almost there, yet still so far to go","volume":"7","author":"Feig","year":"2017","journal-title":"Wiley Interdiscipl. Rev. Comput. Mol. Sci"},{"key":"2023013108053317100_btz101-B14","doi-asserted-by":"crossref","first-page":"282","DOI":"10.1002\/prot.24871","article-title":"Protein structure refinement via molecular-dynamics simulations: what works and what does not?","volume":"84","author":"Feig","year":"2016","journal-title":"Proteins Struct. Funct. Bioinf"},{"key":"2023013108053317100_btz101-B15","doi-asserted-by":"crossref","first-page":"11476","DOI":"10.1038\/srep11476","article-title":"Improving prediction of secondary structure, local backbone angles, and solvent accessible surface area of proteins by iterative deep learning","volume":"5","author":"Heffernan","year":"2015","journal-title":"Sci. Rep"},{"key":"2023013108053317100_btz101-B16","doi-asserted-by":"crossref","first-page":"177","DOI":"10.1002\/prot.25393","article-title":"What makes it difficult to refine protein models further via molecular dynamics simulations?","volume":"86","author":"Heo","year":"2018","journal-title":"Proteins Struct. Funct. Bioinf"},{"key":"2023013108053317100_btz101-B17","doi-asserted-by":"crossref","first-page":"152","DOI":"10.1002\/prot.25409","article-title":"Assessment of the model refinement category in CASP12","volume":"86","author":"Hovan","year":"2018","journal-title":"Proteins Struct. Funct. Bioinf"},{"key":"2023013108053317100_btz101-B18","doi-asserted-by":"crossref","first-page":"327","DOI":"10.1080\/08927020801930539","article-title":"Refinement of comparative models of protein structure by using multicanonical molecular dynamics simulations","volume":"34","author":"Ishitani","year":"2008","journal-title":"Mol. Simul"},{"key":"2023013108053317100_btz101-B19","doi-asserted-by":"crossref","first-page":"2577","DOI":"10.1002\/bip.360221211","article-title":"Dictionary of protein secondary structure: pattern recognition of hydrogen-bonded and geometrical features","volume":"22","author":"Kabsch","year":"1983","journal-title":"Biopolymers"},{"key":"2023013108053317100_btz101-B20","doi-asserted-by":"crossref","first-page":"18949","DOI":"10.1073\/pnas.1115898108","article-title":"Algorithm discovery by protein folding game players","volume":"108","author":"Khatib","year":"2011","journal-title":"Proc. Natl. Acad. Sci. USA"},{"key":"2023013108053317100_btz101-B21","doi-asserted-by":"crossref","first-page":"W526","DOI":"10.1093\/nar\/gkh468","article-title":"Protein structure prediction and analysis using the Robetta server","volume":"32","author":"Kim","year":"2004","journal-title":"Nucleic Acids Res"},{"key":"2023013108053317100_btz101-B22","doi-asserted-by":"crossref","first-page":"321","DOI":"10.1002\/prot.25425","article-title":"Evaluation of the template-based modeling in CASP12","volume":"86","author":"Kryshtafovych","year":"2018","journal-title":"Proteins Struct. Funct. Bioinf"},{"key":"2023013108053317100_btz101-B23","doi-asserted-by":"crossref","first-page":"545","DOI":"10.1016\/B978-0-12-381270-4.00019-6","article-title":"ROSETTA3: an object-oriented software suite for the simulation and design of macromolecules","volume":"487","author":"Leaver-Fay","year":"2011","journal-title":"Methods Enzymol"},{"key":"2023013108053317100_btz101-B24","doi-asserted-by":"crossref","first-page":"293","DOI":"10.1002\/prot.24858","article-title":"Effective protein model structure refinement by loop modeling and overall relaxation","volume":"84","author":"Lee","year":"2016","journal-title":"Proteins Struct. Funct. Bioinf"},{"key":"2023013108053317100_btz101-B25","doi-asserted-by":"crossref","first-page":"709","DOI":"10.1002\/jcc.21664","article-title":"Reliable protein structure refinement using a physical energy function","volume":"32","author":"Lin","year":"2011","journal-title":"J. Computat. Chem"},{"key":"2023013108053317100_btz101-B26","doi-asserted-by":"crossref","first-page":"196","DOI":"10.1002\/prot.24336","article-title":"Physics-based protein structure refinement through multiple molecular dynamics trajectories and structure averaging","volume":"82","author":"Mirjalili","year":"2014","journal-title":"Proteins Struct. Funct. Bioinf"},{"key":"2023013108053317100_btz101-B27","doi-asserted-by":"crossref","first-page":"260","DOI":"10.1002\/prot.25048","article-title":"Assessment of refinement of template-based models in CASP11","volume":"84","author":"Modi","year":"2016","journal-title":"Proteins Struct. Funct. Bioinf"},{"key":"2023013108053317100_btz101-B28","doi-asserted-by":"crossref","first-page":"7","DOI":"10.1002\/prot.25415","article-title":"Critical assessment of methods of protein structure prediction (CASP)\u2014Round XII. Proteins: structure","volume":"86","author":"Moult","year":"2018","journal-title":"Funct. Bioinf"},{"key":"2023013108053317100_btz101-B29","doi-asserted-by":"crossref","first-page":"98","DOI":"10.1002\/prot.24377","article-title":"Evaluation of predictions in the CASP10 model refinement category","volume":"82","author":"Nugent","year":"2014","journal-title":"Proteins Struct. Funct. Bioinf"},{"key":"2023013108053317100_btz101-B30","doi-asserted-by":"crossref","first-page":"314","DOI":"10.1002\/prot.24862","article-title":"CASP11 refinement experiments with ROSETTA","volume":"84","author":"Park","year":"2016","journal-title":"Proteins Struct. Funct. Bioinf"},{"key":"2023013108053317100_btz101-B31","doi-asserted-by":"crossref","first-page":"1974","DOI":"10.1002\/prot.24086","article-title":"Refinement of unreliable local regions in template-based protein models","volume":"80","author":"Park","year":"2012","journal-title":"Proteins Struct. Funct. Bioinf"},{"key":"2023013108053317100_btz101-B32","doi-asserted-by":"crossref","first-page":"2365","DOI":"10.1021\/jp209657n","article-title":"Structure refinement of protein low resolution models using the GNEIMO constrained dynamics method","volume":"116","author":"Park","year":"2012","journal-title":"J. Phys. Chem. B"},{"key":"2023013108053317100_btz101-B33","doi-asserted-by":"crossref","first-page":"2071","DOI":"10.1002\/prot.24098","article-title":"Refinement of protein structure homology models via long, all-atom molecular dynamics simulations","volume":"80","author":"Raval","year":"2012","journal-title":"Proteins Struct. Funct. Bioinf"},{"key":"2023013108053317100_btz101-B34","doi-asserted-by":"crossref","first-page":"W323","DOI":"10.1093\/nar\/gks376","article-title":"KoBaMIN: a knowledge-based minimization web server for protein structure refinement","volume":"40","author":"Rodrigues","year":"2012","journal-title":"Nucleic Acids Res"},{"key":"2023013108053317100_btz101-B35","doi-asserted-by":"crossref","first-page":"66","DOI":"10.1016\/S0076-6879(04)83004-0","article-title":"Protein structure prediction using Rosetta","volume":"383","author":"Rohl","year":"2004","journal-title":"Methods Enzymol"},{"key":"2023013108053317100_btz101-B36","doi-asserted-by":"crossref","first-page":"725.","DOI":"10.1038\/nprot.2010.5","article-title":"I-TASSER: a unified platform for automated protein structure and function prediction","volume":"5","author":"Roy","year":"2010","journal-title":"Nat. Protoc"},{"key":"2023013108053317100_btz101-B37","doi-asserted-by":"crossref","first-page":"3177","DOI":"10.1073\/pnas.0611593104","article-title":"Near-native structure refinement using in vacuo energy minimization","volume":"104","author":"Summa","year":"2007","journal-title":"Proc. Natl. Acad. Sci. USA"},{"key":"2023013108053317100_btz101-B38","doi-asserted-by":"crossref","first-page":"607","DOI":"10.1016\/j.jmb.2010.11.008","article-title":"Alternate states of proteins revealed by detailed energy landscape mapping","volume":"405","author":"Tyka","year":"2011","journal-title":"J. Mol. Biol"},{"key":"2023013108053317100_btz101-B39","doi-asserted-by":"crossref","first-page":"18962","DOI":"10.1038\/srep18962","article-title":"Protein secondary structure prediction using deep convolutional neural fields","volume":"6","author":"Wang","year":"2016","journal-title":"Sci. Rep"},{"key":"2023013108053317100_btz101-B40","first-page":"1","author":"Wang","year":"2016"},{"key":"2023013108053317100_btz101-B41","doi-asserted-by":"crossref","first-page":"5578","DOI":"10.1021\/acs.jctc.5b00942","article-title":"Coupling an ensemble of homologues improves refinement of protein homology models","volume":"11","author":"Wildberg","year":"2015","journal-title":"J. Chem. Theory Comput"},{"key":"2023013108053317100_btz101-B42","doi-asserted-by":"crossref","first-page":"2525","DOI":"10.1016\/j.bpj.2011.10.024","article-title":"Improving the physical realism and structural accuracy of protein models by a two-step atomic-level energy minimization","volume":"101","author":"Xu","year":"2011","journal-title":"Biophys. J"},{"key":"2023013108053317100_btz101-B43","doi-asserted-by":"crossref","first-page":"3370","DOI":"10.1093\/nar\/gkg571","article-title":"LGA: a method for finding 3D similarities in protein structures","volume":"31","author":"Zemla","year":"2003","journal-title":"Nucleic Acids Res"},{"key":"2023013108053317100_btz101-B44","doi-asserted-by":"crossref","first-page":"1784","DOI":"10.1016\/j.str.2011.09.022","article-title":"Atomic-level protein structure refinement using fragment-guided molecular dynamics conformation sampling","volume":"19","author":"Zhang","year":"2011","journal-title":"Structure"},{"key":"2023013108053317100_btz101-B45","doi-asserted-by":"crossref","first-page":"108","DOI":"10.1002\/prot.21702","article-title":"Template-based modeling and free modeling by I-TASSER in CASP7","volume":"69","author":"Zhang","year":"2007","journal-title":"Proteins Struct. Funct. Bioinf"},{"key":"2023013108053317100_btz101-B46","doi-asserted-by":"crossref","first-page":"40.","DOI":"10.1186\/1471-2105-9-40","article-title":"I-TASSER server for protein 3D structure prediction","volume":"9","author":"Zhang","year":"2008","journal-title":"BMC Bioinformatics"}],"container-title":["Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/35\/18\/3320\/48975339\/bioinformatics_35_18_3320.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/35\/18\/3320\/48975339\/bioinformatics_35_18_3320.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,7,14]],"date-time":"2024-07-14T19:34:52Z","timestamp":1720985692000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article\/35\/18\/3320\/5317159"}},"subtitle":[],"editor":[{"given":"Alfonso","family":"Valencia","sequence":"additional","affiliation":[],"role":[{"role":"editor","vocabulary":"crossref"}]}],"short-title":[],"issued":{"date-parts":[[2019,2,13]]},"references-count":46,"journal-issue":{"issue":"18","published-print":{"date-parts":[[2019,9,15]]}},"URL":"https:\/\/doi.org\/10.1093\/bioinformatics\/btz101","relation":{},"ISSN":["1367-4803","1367-4811"],"issn-type":[{"value":"1367-4803","type":"print"},{"value":"1367-4811","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2019,9,15]]},"published":{"date-parts":[[2019,2,13]]}}}