{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,12]],"date-time":"2026-09-12T11:55:31Z","timestamp":1789214131451,"version":"build-2803163510"},"reference-count":52,"publisher":"American Chemical Society (ACS)","issue":"16","license":[{"start":{"date-parts":[[2026,8,3]],"date-time":"2026-08-03T00:00:00Z","timestamp":1785715200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,8,3]],"date-time":"2026-08-03T00:00:00Z","timestamp":1785715200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"},{"start":{"date-parts":[[2026,8,3]],"date-time":"2026-08-03T00:00:00Z","timestamp":1785715200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-045"}],"funder":[{"DOI":"10.13039\/501100010418","name":"Institute for Information and Communications Technology Promotion","doi-asserted-by":"publisher","award":["RS-2023-00220628"],"award-info":[{"award-number":["RS-2023-00220628"]}],"id":[{"id":"10.13039\/501100010418","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,8,24]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Virtual screening (VS) is an essential tool in drug discovery to prioritize potential drug candidates from vast chemical space. One key challenge limiting its performance is accounting for protein conformational flexibility. While ensemble docking methods have been developed to address this challenge by incorporating multiple protein conformations, these methods often rely on computationally intensive physics-based simulations to sample the relevant conformational space. Generative machine learning models offer a highly promising, scalable, and high-throughput alternative to overcome the limitations of these traditional approaches. We therefore investigate whether conformational ensembles generated by BioEmu, a recently developed generative model, can improve VS performance for kinase targets. Using the DUD-E benchmark data set and a validated AutoDock-GPU protocol, we generated and analyzed nearly 1300 structures across 26 kinases (approximately 50 structures each). BioEmu produces structurally diverse ensembles with substantial performance variation among individual structures. However, ensemble methods employing consensus or best-score selection fail to improve upon, and often degrade, VS performance compared to crystal structure baselines. To investigate the source of this limitation, we quantified the relationship between KinCoRe-based conformational state classification and screening performance. By calculating the coefficient of determination (R2) across the kinase subset, we found that the structural features governing VS performance differ substantially from those defining standard conformational states, with KinCoRe classifications leaving over 84% of performance variance unexplained. This critical gap demonstrates that structural diversity alone is insufficient to guarantee screening success. We show that prospective structure selection, rather than structure generation, represents the primary bottleneck in ensemble-based VS, highlighting an urgent need for novel structural descriptors to identify high-performing conformations.<\/jats:p>","DOI":"10.1021\/acs.jcim.6c01677","type":"journal-article","created":{"date-parts":[[2026,8,4]],"date-time":"2026-08-04T14:36:04Z","timestamp":1785854164000},"page":"9905-9915","source":"Crossref","is-referenced-by-count":1,"title":["Evaluating BioEmu-Generated\nKinase Ensembles Reveals\nStructure Selection as the Virtual Screening Bottleneck"],"prefix":"10.1021","volume":"66","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-2365-5793","authenticated-orcid":true,"given":"Jaeoh","family":"Shin","sequence":"first","affiliation":[{"name":"Korea Institute for Advanced Study , , ,","place":["Seoul, Korea, 02455"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Keehyoung","family":"Joo","sequence":"additional","affiliation":[{"name":"Korea Institute for Advanced Study , , ,","place":["Seoul, Korea, 02455"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jejoong","family":"Yoo","sequence":"additional","affiliation":[{"name":"Korea Institute for Advanced Study , , ,","place":["Seoul, Korea, 02455"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"316","published-online":{"date-parts":[[2026,8,3]]},"reference":[{"key":"2026083104122799500_cit1","doi-asserted-by":"publisher","first-page":"343","DOI":"10.3389\/fchem.2020.00343","article-title":"Structure-Based\nVirtual Screening: From Classical to Artificial Intelligence","volume":"8","author":"Maia","year":"2020","journal-title":"Front. Chem."},{"key":"2026083104122799500_cit2","doi-asserted-by":"publisher","first-page":"4112","DOI":"10.1021\/acs.jcim.9b01101","article-title":"Current trends, overlooked\nissues, and unmet challenges\nin virtual screening","volume":"60","author":"Stumpfe","year":"2020","journal-title":"J. Chem. Inf. Model."},{"key":"2026083104122799500_cit3","doi-asserted-by":"publisher","first-page":"673","DOI":"10.1038\/s41586-023-05905-z","article-title":"Computational approaches streamlining\ndrug discovery","volume":"616","author":"Sadybekov","year":"2023","journal-title":"Nature"},{"key":"2026083104122799500_cit4","doi-asserted-by":"publisher","first-page":"102829","DOI":"10.1016\/j.sbi.2024.102829","article-title":"Structure-based virtual screening\nof vast chemical\nspace as a starting point for drug discovery","volume":"87","author":"Carlsson","year":"2024","journal-title":"Curr. Opin. Struct. Biol."},{"key":"2026083104122799500_cit5","doi-asserted-by":"publisher","first-page":"1545","DOI":"10.1021\/cr040423+","article-title":"A statistical thermodynamic\nmodel of the protein ensemble","volume":"106","author":"Hilser","year":"2006","journal-title":"Chem. Rev."},{"key":"2026083104122799500_cit6","doi-asserted-by":"publisher","first-page":"6516","DOI":"10.1021\/acs.chemrev.5b00562","article-title":"Protein ensembles: how does nature\nharness thermodynamic fluctuations for life? The diverse functional\nroles of conformational ensembles in the cell","volume":"116","author":"Wei","year":"2016","journal-title":"Chem. Rev."},{"key":"2026083104122799500_cit7","doi-asserted-by":"publisher","first-page":"850","DOI":"10.1039\/D3CB00114H","article-title":"Protein conformational ensembles\nin function: roles and mechanisms","volume":"4","author":"Nussinov","year":"2023","journal-title":"RSC Chem.\nBiol."},{"key":"2026083104122799500_cit8","doi-asserted-by":"publisher","first-page":"713","DOI":"10.1093\/protein\/12.9.713","article-title":"Folding funnels and\nbinding mechanisms","volume":"12","author":"Ma","year":"1999","journal-title":"Protein Eng."},{"key":"2026083104122799500_cit9","doi-asserted-by":"publisher","first-page":"9970","DOI":"10.1073\/pnas.96.18.9970","article-title":"Folding and binding\ncascades: shifts in energy landscapes","volume":"96","author":"Tsai","year":"1999","journal-title":"Proc.\nNatl. Acad. Sci. U.S.A."},{"key":"2026083104122799500_cit10","doi-asserted-by":"publisher","first-page":"22","DOI":"10.1016\/j.bpc.2013.10.002","article-title":"Multiple conformational selection and induced fit events\ntake place in allosteric propagation","volume":"186","author":"Nussinov","year":"2014","journal-title":"Biophys.\nChem."},{"key":"2026083104122799500_cit11","doi-asserted-by":"publisher","first-page":"2271","DOI":"10.1016\/j.bpj.2018.02.038","article-title":"Ensemble\nDocking\nin Drug Discovery","volume":"114","author":"Amaro","year":"2018","journal-title":"Biophys. J."},{"key":"2026083104122799500_cit12","doi-asserted-by":"publisher","first-page":"060901","DOI":"10.1063\/1.4940794","article-title":"Perspective:\nComputer simulations of long time dynamics","volume":"144","author":"Elber","year":"2016","journal-title":"J. Chem. Phys."},{"key":"2026083104122799500_cit13","doi-asserted-by":"publisher","first-page":"583","DOI":"10.1038\/s41586-021-03819-2","article-title":"Highly accurate protein\nstructure prediction with AlphaFold","volume":"596","author":"Jumper","year":"2021","journal-title":"Nature"},{"key":"2026083104122799500_cit14","doi-asserted-by":"publisher","DOI":"10.7554\/elife.75751","article-title":"Sampling alternative conformational states of transporters\nand receptors with AlphaFold2","volume":"11","author":"del Alamo","year":"2022","journal-title":"eLife"},{"key":"2026083104122799500_cit15","doi-asserted-by":"publisher","first-page":"832","DOI":"10.1038\/s41586-023-06832-9","article-title":"Predicting\nmultiple conformations via sequence clustering and AlphaFold2","volume":"625","author":"Wayment-Steele","year":"2024","journal-title":"Nature"},{"key":"2026083104122799500_cit16","doi-asserted-by":"publisher","first-page":"373","DOI":"10.1038\/s42003-025-07791-9","article-title":"AFsample2 predicts\nmultiple conformations and ensembles\nwith AlphaFold2","volume":"8","author":"Kalakoti","year":"2025","journal-title":"Commun. Biol."},{"key":"2026083104122799500_cit17","doi-asserted-by":"publisher","first-page":"25167","DOI":"10.1038\/s41598-024-75400-6","article-title":"Improving docking\nand virtual screening performance using AlphaFold2 multi-state modeling\nfor kinases","volume":"14","author":"Song","year":"2024","journal-title":"Sci. Rep."},{"key":"2026083104122799500_cit18","doi-asserted-by":"publisher","first-page":"6818","DOI":"10.1073\/pnas.1814279116","article-title":"Defining a new nomenclature for the structures of active\nand inactive kinases","volume":"116","author":"Modi","year":"2019","journal-title":"Proc. Natl. Acad. Sci.\nU.S.A."},{"key":"2026083104122799500_cit19","doi-asserted-by":"publisher","first-page":"6582","DOI":"10.1021\/jm300687e","article-title":"Directory of Useful\nDecoys, Enhanced (DUD-E): Better Ligands and Decoys for Better Benchmarking","volume":"55","author":"Mysinger","year":"2012","journal-title":"J. Med. Chem."},{"key":"2026083104122799500_cit20","doi-asserted-by":"publisher","first-page":"12965","DOI":"10.1038\/ncomms12965","article-title":"Modelling\nproteins\u2019 hidden conformations to predict antibiotic resistance","volume":"7","author":"Hart","year":"2016","journal-title":"Nat. Commun."},{"key":"2026083104122799500_cit21","doi-asserted-by":"publisher","first-page":"RP99702","DOI":"10.7554\/elife.99702","article-title":"Empowering AlphaFold2 for protein\nconformation selective drug discovery with AlphaFold2-RAVE","volume":"13","author":"Gu","year":"2024","journal-title":"eLife"},{"key":"2026083104122799500_cit22","doi-asserted-by":"publisher","first-page":"9214","DOI":"10.1021\/acs.jctc.5c01093","article-title":"Hierarchical AF2RAVE for Multiconformation\nVirtual Screening Targeting S100 Ca2+-Binding Proteins","volume":"21","author":"Gu","year":"2025","journal-title":"J. Chem. Theory Comput."},{"key":"2026083104122799500_cit23","doi-asserted-by":"publisher","first-page":"493","DOI":"10.1038\/s41586-024-07487-w","article-title":"Accurate structure prediction of biomolecular interactions with AlphaFold\n3","volume":"630","author":"Abramson","year":"2024","journal-title":"Nature"},{"key":"2026083104122799500_cit24","doi-asserted-by":"publisher","first-page":"558","DOI":"10.1038\/s42256-024-00837-3","article-title":"Predicting equilibrium\ndistributions for molecular systems with deep\nlearning","volume":"6","author":"Zheng","year":"2024","journal-title":"Nat. Mach. Intell."},{"key":"2026083104122799500_cit25","doi-asserted-by":"publisher","first-page":"arXiv: 2402.04845","DOI":"10.48550\/arXiv.2402.04845","article-title":"AlphaFold Meets Flow\nMatching for\nGenerating Protein Ensembles","author":"Jing","year":"2024","journal-title":"arXiv"},{"key":"2026083104122799500_cit26","doi-asserted-by":"publisher","first-page":"103000","DOI":"10.1016\/j.sbi.2025.103000","article-title":"Modeling Boltzmann-weighted\nstructural ensembles of proteins using artificial intelligence\u2013based\nmethods","volume":"91","author":"Aranganathan","year":"2025","journal-title":"Curr. Opin. Struct. Biol."},{"key":"2026083104122799500_cit27","doi-asserted-by":"publisher","first-page":"103115","DOI":"10.1016\/j.sbi.2025.103115","article-title":"Generation of Protein Dynamics by\nMachine Learning","volume":"93","author":"Janson","year":"2025","journal-title":"Curr. Opin. Struct. Biol."},{"key":"2026083104122799500_cit28","doi-asserted-by":"publisher","DOI":"10.1126\/science.adv9817","article-title":"Scalable emulation of protein equilibrium ensembles\nwith generative deep learning","volume":"389","author":"Lewis","year":"2025","journal-title":"Science"},{"key":"2026083104122799500_cit29","unstructured":"Corso, G.; Sta\u0308rk, H.; Jing, B.; Barzilay, R.; Jaakkola, T.\n          DiffDock: Diffusion\nSteps,\nTwists, and Turns for Molecular Docking. In  International Conference on Learning Representations (ICLR), 2023."},{"key":"2026083104122799500_cit30","doi-asserted-by":"publisher","first-page":"1071","DOI":"10.1038\/s41467-024-45461-2","article-title":"DynamicBind: predicting ligand-specific\nprotein-ligand complex structure\nwith a deep equivariant generative model","volume":"15","author":"Lu","year":"2024","journal-title":"Nat.\nCommun."},{"key":"2026083104122799500_cit31","doi-asserted-by":"publisher","DOI":"10.1126\/science.adl2528","article-title":"Generalized biomolecular\nmodeling and design with RoseTTAFold All-Atom","volume":"384","author":"Krishna","year":"2024","journal-title":"Science"},{"key":"2026083104122799500_cit32","doi-asserted-by":"publisher","first-page":"615955","DOI":"10.1101\/2024.10.10.615955","article-title":"Chai-1. Decoding\nthe molecular interactions\nof life","author":"Boitreaud","year":"2024","journal-title":"bioRxiv"},{"key":"2026083104122799500_cit33","doi-asserted-by":"publisher","first-page":"624167","DOI":"10.1101\/2024.11.19.624167","article-title":"Boltz-1 Democratizing Biomolecular\nInteraction Modeling","author":"Wohlwend","year":"2024","journal-title":"bioRxiv"},{"key":"2026083104122799500_cit34","doi-asserted-by":"publisher","first-page":"11","DOI":"10.3389\/fphar.2018.00011","article-title":"Decoys selection\nin benchmarking datasets: overview and perspectives","volume":"9","author":"Re\u0301au","year":"2018","journal-title":"Front. Pharmacol."},{"key":"2026083104122799500_cit35","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0220113","article-title":"Hidden bias in the\nDUD-E dataset leads to misleading performance of deep learning in\nstructure-based virtual screening","volume":"14","author":"Chen","year":"2019","journal-title":"PLoS One"},{"key":"2026083104122799500_cit36","doi-asserted-by":"publisher","first-page":"19","DOI":"10.1016\/j.softx.2015.06.001","article-title":"GROMACS: High\nperformance\nmolecular simulations through multi-level parallelism from laptops\nto supercomputers","volume":"1","author":"Abraham","year":"2015","journal-title":"SoftwareX"},{"key":"2026083104122799500_cit37","doi-asserted-by":"publisher","first-page":"1950","DOI":"10.1002\/prot.22711","article-title":"Improved side-chain\ntorsion potentials for the Amber ff99SB protein force field","volume":"78","author":"Lindorff-Larsen","year":"2010","journal-title":"Proteins: Struct., Funct., Bioinf."},{"key":"2026083104122799500_cit38","doi-asserted-by":"publisher","first-page":"btaf010","DOI":"10.1093\/bioinformatics\/btaf010","article-title":"FlowPacker: protein side-chain packing with torsional\nflow matching","volume":"41","author":"Lee","year":"2025","journal-title":"Bioinformatics"},{"key":"2026083104122799500_cit39","doi-asserted-by":"publisher","first-page":"1060","DOI":"10.1021\/acs.jctc.0c01006","article-title":"Accelerating\nAutoDock4 with GPUs and Gradient-Based Local Search","volume":"17","author":"Santos-Martins","year":"2021","journal-title":"J. Chem. Theory Comput."},{"key":"2026083104122799500_cit40","unstructured":"The PyMOL Molecular\nGraphics System, Version 2.5.0 Schro\u0308dinger,\nLLC."},{"key":"2026083104122799500_cit41","doi-asserted-by":"publisher","first-page":"1982","DOI":"10.1021\/acs.jcim.2c01504","article-title":"Vina-GPU 2.0: further accelerating\nAutoDock Vina and its derivatives with graphics processing units","volume":"63","author":"Ding","year":"2023","journal-title":"J. Chem. Inf. Model."},{"key":"2026083104122799500_cit42","doi-asserted-by":"publisher","first-page":"2382","DOI":"10.1109\/TCBB.2024.3467127","article-title":"Vina-GPU 2.1: towards further optimizing docking speed and precision\nof AutoDock Vina and its derivatives","volume":"21","author":"Tang","year":"2024","journal-title":"IEEE\/ACM\nTrans. Comput. Biol. Bioinf."},{"key":"2026083104122799500_cit43","doi-asserted-by":"publisher","first-page":"293","DOI":"10.1002\/pro.3330","article-title":"MolProbity: more and better reference data for improved\nall-atom structure validation","volume":"27","author":"Williams","year":"2018","journal-title":"Protein Sci."},{"key":"2026083104122799500_cit44","doi-asserted-by":"publisher","first-page":"702","DOI":"10.1002\/prot.20264","article-title":"Scoring function for automated assessment of protein\nstructure template quality","volume":"57","author":"Zhang","year":"2004","journal-title":"Proteins: Struct.,\nFunct., Bioinf."},{"key":"2026083104122799500_cit45","doi-asserted-by":"publisher","first-page":"D562","DOI":"10.1093\/nar\/gkaa895","article-title":"KLIFS:\nan overhaul after the first 5 years of supporting kinase research","volume":"49","author":"Kanev","year":"2021","journal-title":"Nucleic Acids Res."},{"key":"2026083104122799500_cit46","doi-asserted-by":"publisher","first-page":"488","DOI":"10.1021\/ci600426e","article-title":"Evaluating virtual screening methods: good and bad\nmetrics for the \u201cearly recognition\u201d problem","volume":"47","author":"Truchon","year":"2007","journal-title":"J. Chem. Inf. Model."},{"key":"2026083104122799500_cit47","doi-asserted-by":"publisher","first-page":"W463","DOI":"10.1093\/nar\/gkaf361","article-title":"PLIP 2025:\nintroducing protein\u2013protein interactions to the protein\u2013ligand\ninteraction profiler","volume":"53","author":"Schake","year":"2025","journal-title":"Nucleic Acids Res."},{"key":"2026083104122799500_cit48","doi-asserted-by":"publisher","first-page":"7168","DOI":"10.1021\/acs.jcim.6c01000","article-title":"Accelerated Sampling of Protein Dynamics\nUsing BioEmu-Augmented\nMolecular Simulation","volume":"66","author":"Bhakat","year":"2026","journal-title":"J. Chem. Inf. Model."},{"key":"2026083104122799500_cit49","doi-asserted-by":"publisher","first-page":"1181","DOI":"10.1110\/ps.8.6.1181","article-title":"Folding funnels,\nbinding\nfunnels, and protein function","volume":"8","author":"Tsai","year":"1999","journal-title":"Protein Sci."},{"key":"2026083104122799500_cit50","doi-asserted-by":"publisher","first-page":"418","DOI":"10.1002\/prot.1107","article-title":"Structured disorder\nand conformational selection","volume":"44","author":"Tsai","year":"2001","journal-title":"Proteins: Struct.,\nFunct., Bioinf."},{"key":"2026083104122799500_cit51","doi-asserted-by":"publisher","first-page":"4263","DOI":"10.1021\/acs.jcim.0c00155","article-title":"LIT-PCBA: an unbiased data set for\nmachine learning and virtual screening","volume":"60","author":"Tran-Nguyen","year":"2020","journal-title":"J. Chem.\nInf. Model."},{"key":"2026083104122799500_cit52","doi-asserted-by":"publisher","first-page":"10167","DOI":"10.1021\/acs.jcim.5c01765","article-title":"Sampling and Ranking of Protein Conformations\nUsing Machine Learning Techniques Do Not Improve the Quality of Rigid\nProtein\u2013Protein Docking","volume":"65","author":"Stratiichuk","year":"2025","journal-title":"J. Chem. Inf.\nModel."}],"container-title":["Journal of Chemical Information\nand Modeling"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/pubs.acs.org\/doi\/pdf\/10.1021\/acs.jcim.6c01677","content-type":"application\/pdf","content-version":"vor","intended-application":"unspecified"},{"URL":"https:\/\/pubs.acs.org\/jcisd8\/article-pdf\/66\/16\/9905\/66377205\/acs.jcim.6c01677.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/pubs.acs.org\/jcisd8\/article-pdf\/66\/16\/9905\/66377205\/acs.jcim.6c01677.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,31]],"date-time":"2026-08-31T08:12:35Z","timestamp":1788163955000},"score":1,"resource":{"primary":{"URL":"https:\/\/pubs.acs.org\/jcisd8\/article\/66\/16\/9905\/5242069\/Evaluating-BioEmu-Generated-Kinase-Ensembles"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,3]]},"references-count":52,"journal-issue":{"issue":"16","published-online":{"date-parts":[[2026,8,3]]},"published-print":{"date-parts":[[2026,8,24]]}},"URL":"https:\/\/doi.org\/10.1021\/acs.jcim.6c01677","relation":{},"ISSN":["1549-9596","1549-960X"],"issn-type":[{"value":"1549-9596","type":"print"},{"value":"1549-960X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,3]]}}}