{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,2]],"date-time":"2026-06-02T02:20:03Z","timestamp":1780366803704,"version":"3.54.1"},"reference-count":44,"publisher":"Oxford University Press (OUP)","issue":"5","license":[{"start":{"date-parts":[[2017,10,12]],"date-time":"2017-10-12T00:00:00Z","timestamp":1507766400000},"content-version":"vor","delay-in-days":1,"URL":"http:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100003382","name":"Core Research for Evolutional Science and Technology","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100003382","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100003382","name":"CREST","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100003382","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001695","name":"JST","doi-asserted-by":"publisher","award":["JPMJCR1502"],"award-info":[{"award-number":["JPMJCR1502"]}],"id":[{"id":"10.13039\/501100001695","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001695","name":"JST","doi-asserted-by":"publisher","award":["JPMJCR1311"],"award-info":[{"award-number":["JPMJCR1311"]}],"id":[{"id":"10.13039\/501100001695","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2018,3,1]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:sec><jats:title>Motivation<\/jats:title><jats:p>Fast and accurate prediction of protein\u2013ligand binding structures is indispensable for structure-based drug design and accurate estimation of binding free energy of drug candidate molecules in drug discovery. Recently, accurate pose prediction methods based on short Molecular Dynamics (MD) simulations, such as MM-PBSA and MM-GBSA, among generated docking poses have been used. Since molecular structures obtained from MD simulation depend on the initial condition, taking the average over different initial conditions leads to better accuracy. Prediction accuracy of protein\u2013ligand binding poses can be improved with multiple runs at different initial velocity.<\/jats:p><\/jats:sec><jats:sec><jats:title>Results<\/jats:title><jats:p>This paper shows that a machine learning method, called Best Arm Identification, can optimally control the number of MD runs for each binding pose. It allows us to identify a correct binding pose with a minimum number of total runs. Our experiment using three proteins and eight inhibitors showed that the computational cost can be reduced substantially without sacrificing accuracy. This method can be applied for controlling all kinds of molecular simulations to obtain best results under restricted computational resources.<\/jats:p><\/jats:sec><jats:sec><jats:title>Availability and implementation<\/jats:title><jats:p>Code and data are available on GitHub at https:\/\/github.com\/tsudalab\/bpbi.<\/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\/btx638","type":"journal-article","created":{"date-parts":[[2017,10,10]],"date-time":"2017-10-10T11:10:27Z","timestamp":1507633827000},"page":"770-778","source":"Crossref","is-referenced-by-count":29,"title":["Machine learning accelerates MD-based binding pose prediction between ligands and proteins"],"prefix":"10.1093","volume":"34","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-3914-248X","authenticated-orcid":false,"given":"Kei","family":"Terayama","sequence":"first","affiliation":[{"name":"Department of Computational Biology and Medical Science, Graduate School of Frontier Sciences, The University of Tokyo, Chiba, Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hiroaki","family":"Iwata","sequence":"additional","affiliation":[{"name":"Foundation for Biomedical Research and Innovation, Hyogo, Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mitsugu","family":"Araki","sequence":"additional","affiliation":[{"name":"RIKEN Advanced Institute for Computational Science, Hyogo, Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yasushi","family":"Okuno","sequence":"additional","affiliation":[{"name":"RIKEN Advanced Institute for Computational Science, Hyogo, Japan"},{"name":"Department of Biomedical Data Intelligence, Graduate School of Medicine, Kyoto University, Kyoto, Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Koji","family":"Tsuda","sequence":"additional","affiliation":[{"name":"Department of Computational Biology and Medical Science, Graduate School of Frontier Sciences, The University of Tokyo, Chiba, Japan"},{"name":"Center for Materials Research by Information Integration, NIMS, Ibaraki, Japan"},{"name":"RIKEN Center for Advanced Intelligence Project, Tokyo, Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"286","published-online":{"date-parts":[[2017,10,11]]},"reference":[{"key":"2023012712412447900_btx638-B1","first-page":"39.1","author":"Agrawal","year":"2012"},{"key":"2023012712412447900_btx638-B2","doi-asserted-by":"crossref","first-page":"385","DOI":"10.1093\/protein\/7.3.385","article-title":"A new method for predicting binding affinity in computer-aided drug design","volume":"7","author":"\u00c5qvist","year":"1994","journal-title":"Protein Eng"},{"key":"2023012712412447900_btx638-B3","author":"Audibert","year":"2010"},{"key":"2023012712412447900_btx638-B4","doi-asserted-by":"crossref","first-page":"235","DOI":"10.1023\/A:1013689704352","article-title":"Finite-time analysis of the multiarmed bandit problem","volume":"47","author":"Auer","year":"2002","journal-title":"Mach. Learn"},{"key":"2023012712412447900_btx638-B5","doi-asserted-by":"crossref","first-page":"1542","DOI":"10.1002\/prot.24302","article-title":"The stability of cylindrin \u03b2-barrel amyloid oligomer models\u2014a molecular dynamics study","volume":"81","author":"Berhanu","year":"2013","journal-title":"Proteins Struct. Funct. Bioinf"},{"key":"2023012712412447900_btx638-B6","doi-asserted-by":"crossref","first-page":"235","DOI":"10.1093\/nar\/28.1.235","article-title":"The protein data bank","volume":"28","author":"Berman","year":"2000","journal-title":"Nucleic Acids Res"},{"key":"2023012712412447900_btx638-B7","author":"Bubeck","year":"2009"},{"key":"2023012712412447900_btx638-B8","volume-title":"Amber 12","author":"Case","year":"2012"},{"key":"2023012712412447900_btx638-B9","doi-asserted-by":"crossref","first-page":"1079","DOI":"10.1021\/ci9000053","article-title":"Comparative assessment of scoring functions on a diverse test set","volume":"49","author":"Cheng","year":"2009","journal-title":"J. Chem. Inf. Model"},{"key":"2023012712412447900_btx638-B10","doi-asserted-by":"crossref","first-page":"7361","DOI":"10.1021\/ja100259r","article-title":"Single-molecule pulling simulations can discern active from inactive enzyme inhibitors","volume":"132","author":"Colizzi","year":"2010","journal-title":"J. Am. Chem. Soc"},{"key":"2023012712412447900_btx638-B11","first-page":"72","author":"Coulom","year":"2006"},{"key":"2023012712412447900_btx638-B12","doi-asserted-by":"crossref","first-page":"10089","DOI":"10.1063\/1.464397","article-title":"Particle mesh Ewald: an N log (N) method for Ewald sums in large systems","volume":"98","author":"Darden","year":"1993","journal-title":"J. Chem. Phys"},{"key":"2023012712412447900_btx638-B13","doi-asserted-by":"crossref","first-page":"021914.","DOI":"10.1103\/PhysRevE.79.021914","article-title":"Massively parallel computation of absolute binding free energy with well-equilibrated states","volume":"79","author":"Fujitani","year":"2009","journal-title":"Phys. Rev. E"},{"key":"2023012712412447900_btx638-B14","first-page":"2222","author":"Gabillon","year":"2011"},{"key":"2023012712412447900_btx638-B15","author":"Gabillon","year":"2012"},{"key":"2023012712412447900_btx638-B16","doi-asserted-by":"crossref","first-page":"837","DOI":"10.1002\/jcc.21366","article-title":"How to obtain statistically converged MM\/GBSA results","volume":"31","author":"Genheden","year":"2010","journal-title":"J. Comput. Chem"},{"key":"2023012712412447900_btx638-B17","doi-asserted-by":"crossref","first-page":"116","DOI":"10.1021\/ct700200b","article-title":"P-LINCS: a parallel linear constraint solver for molecular simulation","volume":"4","author":"Hess","year":"2008","journal-title":"J. Chem. Theory Comput"},{"key":"2023012712412447900_btx638-B18","doi-asserted-by":"crossref","first-page":"435","DOI":"10.1021\/ct700301q","article-title":"GROMACS 4: algorithms for highly efficient, load-balanced, and scalable molecular simulation","volume":"4","author":"Hess","year":"2008","journal-title":"J. Chem. Theory Comput"},{"key":"2023012712412447900_btx638-B19","doi-asserted-by":"crossref","first-page":"1177","DOI":"10.1021\/jm0609162","article-title":"Molecular dynamics and free energy studies on the wild-type and double mutant HIV-1 protease complexed with amprenavir and two amprenavir-related inhibitors: mechanism for binding and drug resistance","volume":"50","author":"Hou","year":"2007","journal-title":"J. Med. Chem"},{"key":"2023012712412447900_btx638-B20","doi-asserted-by":"crossref","first-page":"69","DOI":"10.1021\/ci100275a","article-title":"Assessing the performance of the MM\/PBSA and MM\/GBSA methods. 1. The accuracy of binding free energy calculations based on molecular dynamics simulations","volume":"51","author":"Hou","year":"2010","journal-title":"J. Chem. Inf. Model"},{"key":"2023012712412447900_btx638-B21","doi-asserted-by":"crossref","first-page":"866","DOI":"10.1002\/jcc.21666","article-title":"Assessing the performance of the molecular mechanics\/Poisson Boltzmann surface area and molecular mechanics\/generalized born surface area methods. II. The accuracy of ranking poses generated from docking","volume":"32","author":"Hou","year":"2011","journal-title":"J. Comput. Chem"},{"key":"2023012712412447900_btx638-B22","doi-asserted-by":"crossref","first-page":"926","DOI":"10.1063\/1.445869","article-title":"Comparison of simple potential functions for simulating liquid water","volume":"79","author":"Jorgensen","year":"1983","journal-title":"J. Chem. Phys"},{"key":"2023012712412447900_btx638-B23","author":"Kaufmann","year":"2013"},{"key":"2023012712412447900_btx638-B24","doi-asserted-by":"crossref","first-page":"889","DOI":"10.1021\/ar000033j","article-title":"Calculating structures and free energies of complex molecules: combining molecular mechanics and continuum models","volume":"33","author":"Kollman","year":"2000","journal-title":"Acc. Chem. Res"},{"key":"2023012712412447900_btx638-B25","doi-asserted-by":"crossref","first-page":"1951","DOI":"10.1021\/ci500020m","article-title":"g_mmpbsa\u2014a GROMACS tool for high-throughput MM-PBSA calculations","volume":"54","author":"Kumari","year":"2014","journal-title":"J. Chem. Inf. Model"},{"key":"2023012712412447900_btx638-B26","doi-asserted-by":"crossref","first-page":"2839","DOI":"10.2174\/09298673113209990001","article-title":"Virtual screening strategies in drug discovery: a critical review","volume":"20","author":"Lavecchia","year":"2013","journal-title":"Curr. Med. Chem"},{"key":"2023012712412447900_btx638-B27","first-page":"661","author":"Li","year":"2010"},{"key":"2023012712412447900_btx638-B28","doi-asserted-by":"crossref","first-page":"1950","DOI":"10.1002\/prot.22711","article-title":"Improved side-chain torsion potentials for the Amber ff99SB protein force field","volume":"78","author":"Lindorff-Larsen","year":"2010","journal-title":"Proteins Struct. Funct. Bioinf"},{"key":"2023012712412447900_btx638-B29","doi-asserted-by":"crossref","first-page":"1923","DOI":"10.2174\/1568026614666140929124445","article-title":"Structure-based virtual screening for drug discovery: principles, applications and recent advances","volume":"14","author":"Lionta","year":"2014","journal-title":"Curr. Top. Med. Chem"},{"key":"2023012712412447900_btx638-B30","doi-asserted-by":"crossref","first-page":"527","DOI":"10.1007\/s10822-011-9524-z","article-title":"Binding affinities in the SAMPL3 trypsin and host\u2013guest blind tests estimated with the MM\/PBSA and LIE methods","volume":"26","author":"Mikulskis","year":"2012","journal-title":"J. Comput. Aided Mol. Des"},{"key":"2023012712412447900_btx638-B31","doi-asserted-by":"crossref","first-page":"3314","DOI":"10.1021\/ct300418h","article-title":"MMPBSA.py: an efficient program for end-state free energy calculations","volume":"8","author":"Miller","year":"2012","journal-title":"J. Chem. Theory Comput"},{"key":"2023012712412447900_btx638-B32","doi-asserted-by":"crossref","first-page":"e1000528.","DOI":"10.1371\/journal.pcbi.1000528","article-title":"High-performance drug discovery: computational screening by combining docking and molecular dynamics simulations","volume":"5","author":"Okimoto","year":"2009","journal-title":"PLoS Comput. Biol"},{"key":"2023012712412447900_btx638-B33","doi-asserted-by":"crossref","first-page":"3712","DOI":"10.1021\/jp994072s","article-title":"Modification of the generalized born model suitable for macromolecules","volume":"104","author":"Onufriev","year":"2000","journal-title":"J. Phys. Chem. B"},{"key":"2023012712412447900_btx638-B34","doi-asserted-by":"crossref","first-page":"144","DOI":"10.1016\/j.bpj.2011.11.4008","article-title":"Discrete molecular dynamics distinguishes nativelike binding poses from decoys in difficult targets","volume":"102","author":"Proctor","year":"2012","journal-title":"Biophys. J"},{"key":"2023012712412447900_btx638-B35","doi-asserted-by":"crossref","first-page":"527","DOI":"10.1090\/S0002-9904-1952-09620-8","article-title":"Some aspects of the sequential design of experiments","volume":"58","author":"Robbins","year":"1952","journal-title":"Bull. Amer. Math. Soc"},{"key":"2023012712412447900_btx638-B36","doi-asserted-by":"crossref","first-page":"890","DOI":"10.1021\/ci100007w","article-title":"Accurate ensemble molecular dynamics binding free energy ranking of multidrug-resistant HIV-1 proteases","volume":"50","author":"Sadiq","year":"2010","journal-title":"J. Chem. Inf. Model"},{"key":"2023012712412447900_btx638-B37","doi-asserted-by":"crossref","first-page":"148","DOI":"10.1109\/JPROC.2015.2494218","article-title":"Taking the human out of the loop: a review of Bayesian optimization","volume":"104","author":"Shahriari","year":"2016","journal-title":"Proc. IEEE"},{"key":"2023012712412447900_btx638-B38","doi-asserted-by":"crossref","first-page":"484","DOI":"10.1038\/nature16961","article-title":"Mastering the game of go with deep neural networks and tree search","volume":"529","author":"Silver","year":"2016","journal-title":"Nature"},{"key":"2023012712412447900_btx638-B39","doi-asserted-by":"crossref","first-page":"9401","DOI":"10.1021\/ja981844+","article-title":"Continuum solvent studies of the stability of DNA, RNA, and phosphoramidate-DNA helices","volume":"120","author":"Srinivasan","year":"1998","journal-title":"J. Am. Chem. Soc"},{"key":"2023012712412447900_btx638-B40","doi-asserted-by":"crossref","first-page":"1081","DOI":"10.1021\/ci700470c","article-title":"Investigation of MM-PBSA rescoring of docking poses","volume":"48","author":"Thompson","year":"2008","journal-title":"J. Chem. Inf. Model"},{"key":"2023012712412447900_btx638-B41","doi-asserted-by":"crossref","first-page":"199.","DOI":"10.1214\/14-STS504","article-title":"Multi-armed bandit models for the optimal design of clinical trials: benefits and challenges","volume":"30","author":"Villar","year":"2015","journal-title":"Stat. Sci"},{"key":"2023012712412447900_btx638-B42","doi-asserted-by":"crossref","first-page":"1157","DOI":"10.1002\/jcc.20035","article-title":"Development and testing of a general amber force field","volume":"25","author":"Wang","year":"2004","journal-title":"J. Comput. Chem"},{"key":"2023012712412447900_btx638-B43","doi-asserted-by":"crossref","first-page":"8408","DOI":"10.1021\/jp404160y","article-title":"Assessing the performance of MM\/PBSA and MM\/GBSA methods. 3. The impact of force fields and ligand charge models","volume":"117","author":"Xu","year":"2013","journal-title":"J. Phys. Chem. B"},{"key":"2023012712412447900_btx638-B44","doi-asserted-by":"crossref","first-page":"1940","DOI":"10.1002\/prot.23018","article-title":"Virtual screening using molecular simulations","volume":"79","author":"Yang","year":"2011","journal-title":"Proteins Struct. Funct. Bioinf"}],"container-title":["Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/34\/5\/770\/48914781\/bioinformatics_34_5_770.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/34\/5\/770\/48914781\/bioinformatics_34_5_770.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,26]],"date-time":"2025-06-26T05:19:50Z","timestamp":1750915190000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article\/34\/5\/770\/4457357"}},"subtitle":[],"editor":[{"given":"Burkhard","family":"Rost","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2017,10,11]]},"references-count":44,"journal-issue":{"issue":"5","published-print":{"date-parts":[[2018,3,1]]}},"URL":"https:\/\/doi.org\/10.1093\/bioinformatics\/btx638","relation":{},"ISSN":["1367-4803","1367-4811"],"issn-type":[{"value":"1367-4803","type":"print"},{"value":"1367-4811","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2018,3,1]]},"published":{"date-parts":[[2017,10,11]]}}}