{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,14]],"date-time":"2026-09-14T14:18:48Z","timestamp":1789395528900,"version":"build-2803163510"},"reference-count":44,"publisher":"American Chemical Society (ACS)","issue":"17","license":[{"start":{"date-parts":[[2026,8,20]],"date-time":"2026-08-20T00:00:00Z","timestamp":1787184000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,8,20]],"date-time":"2026-08-20T00:00:00Z","timestamp":1787184000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"},{"start":{"date-parts":[[2026,8,20]],"date-time":"2026-08-20T00:00:00Z","timestamp":1787184000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-045"}],"funder":[{"DOI":"10.13039\/100006942","name":"Medical University of South Carolina","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100006942","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001343","name":"University of Pretoria","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100001343","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Boltz-2 team"},{"name":"Avicenna Biosciences Inc"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,9,14]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>The rapid advancement of machine learning (ML)-based protein structure prediction, exemplified by AlphaFold2 and extended by newer models such as AlphaFold3 and Boltz-2, has generated significant optimism for structure-guided drug discovery. In particular, ligand\u2013protein cofolding approaches offer the potential to overcome limitations in generating starting structures for physics-based free energy perturbation (FEP) calculations. However, the practical readiness of ML-predicted structures for FEP applications remains insufficiently evaluated. Here, we systematically assess experimentally determined crystal structures, a homology model, and ML-predicted protein structures as inputs for FEP using a well-characterized congeneric series targeting the tyrosine kinase cSrc. A data set of 133 compounds was evaluated through more than 1400 FEP calculations under minimal optimization to approximate \u201cout-of-the-box\u201d performance. By maintaining consistent preparation protocols, we isolate the impact of structural origin on predictive accuracy. Variable performance was observed across both experimental and ML-predicted structures, highlighting that even under this idealized benchmark scenario, significant challenges remain in reliably generating and refining predictive protein\u2013ligand complexes. This study demonstrates that predictive variation in micro and macro conformational states\u2500rather than the structural source\u2500governs predictive reliability, underscoring the need for careful validation when integrating ML-derived structures into FEP workflows.<\/jats:p>","DOI":"10.1021\/acs.jcim.6c01024","type":"journal-article","created":{"date-parts":[[2026,8,20]],"date-time":"2026-08-20T07:01:25Z","timestamp":1787209285000},"page":"10992-11005","source":"Crossref","is-referenced-by-count":0,"title":["To ML-Predict or\nNot to ML-Predict: The Impact of\nMachine Learning-Predicted Protein Structures on FEP Accuracy and\nData Augmentation"],"prefix":"10.1021","volume":"66","author":[{"given":"Parker","family":"Dryja","sequence":"first","affiliation":[{"name":"Avicenna Biosciences Inc. , 101 W Chapel Hill St , , ,","place":["Durham, North Carolina, United States, 27001"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Morne\u0301","family":"Muller","sequence":"additional","affiliation":[{"name":"Avicenna Biosciences Inc. , 101 W Chapel Hill St , , ,","place":["Durham, North Carolina, United States, 27001"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Monique","family":"Horn","sequence":"additional","affiliation":[{"name":"University of Pretoria , , ,","place":["Pretoria, South Africa, 0001"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ilya","family":"Balabin","sequence":"additional","affiliation":[{"name":"Avicenna Biosciences Inc. , 101 W Chapel Hill St , , ,","place":["Durham, North Carolina, United States, 27001"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zackery W.","family":"Dentmon","sequence":"additional","affiliation":[{"name":"Avicenna Biosciences Inc. , 101 W Chapel Hill St , , ,","place":["Durham, North Carolina, United States, 27001"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Prawin","family":"Rimal","sequence":"additional","affiliation":[{"name":"Clemson University , , , ,","place":["Clemson, South Carolina, United States, 29634"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4005-9678","authenticated-orcid":true,"given":"Yuri K.","family":"Peterson","sequence":"additional","affiliation":[{"name":"Medical University of South Carolina , , , ,","place":["Charleston, South Carolina, United States, 29425-1410"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Fourie","family":"Joubert","sequence":"additional","affiliation":[{"name":"University of Pretoria , , ,","place":["Pretoria, South Africa, 0001"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Thomas M.","family":"Kaiser","sequence":"additional","affiliation":[{"name":"Avicenna Biosciences Inc. , 101 W Chapel Hill St , , ,","place":["Durham, North Carolina, United States, 27001"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0272-1111","authenticated-orcid":true,"given":"Pieter B.","family":"Burger","sequence":"additional","affiliation":[{"name":"Avicenna Biosciences Inc. , 101 W Chapel Hill St , , ,","place":["Durham, North Carolina, United States, 27001"]},{"name":"Medical University of South Carolina , , , ,","place":["Charleston, South Carolina, United States, 29425-1410"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"316","published-online":{"date-parts":[[2026,8,20]]},"reference":[{"key":"2026091409073413400_cit1","first-page":"203","article-title":"It will change\neverything\u2019: DeepMind\u2019s AI makes gigantic leap in solving\nprotein structures","author":"Callaway","journal-title":"Nature"},{"issue":"8016","key":"2026091409073413400_cit2","doi-asserted-by":"publisher","first-page":"493","DOI":"10.1038\/s41586-024-07487-w","article-title":"Accurate structure prediction\nof biomolecular interactions with AlphaFold 3","volume":"630","author":"Abramson","year":"2024","journal-title":"Nature"},{"key":"2026091409073413400_cit3","unstructured":"Boitreaud, J.; Dent, J.; McPartlon, M.; et al.\n          Chai-1: Decoding\nthe molecular interactions of life\n          bioRxiv.\n          2024\n          10.1101\/2024.10.10.615955."},{"issue":"7873","key":"2026091409073413400_cit4","doi-asserted-by":"publisher","first-page":"583","DOI":"10.1038\/s41586-021-03819-2","article-title":"Highly\naccurate protein\nstructure prediction with AlphaFold","volume":"596","author":"Jumper","year":"2021","journal-title":"Nature"},{"key":"2026091409073413400_cit5","doi-asserted-by":"crossref","unstructured":"Passaro, S.; Corso, G.; Wohlwend, J.; et al.\n          Boltz-2: Towards\nAccurate and Efficient Binding Affinity Prediction\n          bioRxiv.\n          2025\n          10.1101\/2025.06.14.659707.","DOI":"10.1101\/2025.06.14.659707"},{"issue":"18","key":"2026091409073413400_cit6","doi-asserted-by":"publisher","first-page":"4351","DOI":"10.1021\/acs.jcim.2c00796","article-title":"Are Deep Learning Structural Models\nSufficiently Accurate for Free-Energy Calculations? Application of\nFEP+ to AlphaFold2-Predicted Structures","volume":"62","author":"Beuming","year":"2022","journal-title":"J.\nChem. Inf. Model."},{"key":"2026091409073413400_cit7","doi-asserted-by":"publisher","first-page":"950","DOI":"10.1038\/s41589-024-01638-w","article-title":"The power and pitfalls\nof AlphaFold2 for structure\nprediction beyond rigid globular proteins","volume":"20","author":"Agarwal","year":"2024","journal-title":"Nat.\nChem. Biol."},{"issue":"1","key":"2026091409073413400_cit8","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-025-63947-5","article-title":"Investigating whether deep learning\nmodels for co-folding learn the physics of protein-ligand interactions","volume":"16","author":"Masters","year":"2025","journal-title":"Nat. Commun."},{"issue":"34","key":"2026091409073413400_cit9","doi-asserted-by":"publisher","first-page":"6372","DOI":"10.1021\/acs.jpcb.2c04346","article-title":"AlphaFold,\nArtificial\nIntelligence (AI), and Allostery","volume":"126","author":"Nussinov","year":"2022","journal-title":"J. Phys. Chem.\nB"},{"issue":"7","key":"2026091409073413400_cit10","doi-asserted-by":"publisher","first-page":"4262","DOI":"10.1021\/acs.jctc.1c00194","article-title":"PyAutoFEP: An Automated\nFree Energy Perturbation Workflow for GROMACS Integrating Enhanced\nSampling Methods","volume":"17","author":"Carvalho Martins","year":"2021","journal-title":"J. Chem. Theory Comput."},{"issue":"6","key":"2026091409073413400_cit11","doi-asserted-by":"publisher","first-page":"3120","DOI":"10.1021\/acs.jcim.0c00165","article-title":"Assessment of Binding Affinity via Alchemical Free-Energy\nCalculations","volume":"60","author":"Kuhn","year":"2020","journal-title":"J. Chem. Inf. Model."},{"issue":"7","key":"2026091409073413400_cit12","doi-asserted-by":"publisher","first-page":"2695","DOI":"10.1021\/ja512751q","article-title":"Accurate and Reliable Prediction of Relative\nLigand Binding Potency in Prospective Drug Discovery by Way of a Modern\nFree-Energy Calculation Protocol and Force Field","volume":"137","author":"Wang","year":"2015","journal-title":"J. Am. Chem. Soc."},{"issue":"13","key":"2026091409073413400_cit13","doi-asserted-by":"publisher","first-page":"6927","DOI":"10.1021\/acs.jcim.5c00947","article-title":"Quantification of the\nImpact of Structure Quality on Predicted Binding Free Energy Accuracy","volume":"65","author":"Behera","year":"2025","journal-title":"J. Chem. Inf. Model."},{"issue":"3","key":"2026091409073413400_cit14","doi-asserted-by":"publisher","first-page":"521","DOI":"10.1016\/j.cell.2023.12.034","article-title":"Enabling structure-based\ndrug discovery utilizing predicted models","volume":"187","author":"Miller","year":"2024","journal-title":"Cell."},{"issue":"7","key":"2026091409073413400_cit15","doi-asserted-by":"publisher","first-page":"1625","DOI":"10.1021\/acs.accounts.7b00083","article-title":"Advancing\nDrug Discovery through Enhanced Free Energy Calculations","volume":"50","author":"Abel","year":"2017","journal-title":"Acc. Chem. Res."},{"issue":"9","key":"2026091409073413400_cit16","doi-asserted-by":"publisher","first-page":"3812","DOI":"10.1021\/acs.jcim.4c00071","article-title":"FEP Augmentation as a Means to Solve Data\nPaucity Problems for Machine Learning in Chemical Biology","volume":"64","author":"Burger","year":"2024","journal-title":"J. Chem. Inf Model."},{"issue":"12","key":"2026091409073413400_cit17","doi-asserted-by":"publisher","first-page":"2911","DOI":"10.1021\/acs.jcim.7b00564","article-title":"Relative Binding\nFree Energy Calculations\nin Drug Discovery: Recent Advances and Practical Considerations","volume":"57","author":"Cournia","year":"2017","journal-title":"J. Chem. Inf. Model."},{"issue":"1","key":"2026091409073413400_cit18","doi-asserted-by":"publisher","first-page":"222","DOI":"10.1038\/s42004-023-01019-9","article-title":"The maximal and current accuracy of rigorous\nprotein-ligand binding free energy calculations","volume":"6","author":"Ross","year":"2023","journal-title":"Commun. Chem."},{"issue":"11","key":"2026091409073413400_cit19","doi-asserted-by":"publisher","first-page":"5563","DOI":"10.1021\/acs.jcim.0c00449","article-title":"Accurate Prediction\nof GPCR Ligand Binding Affinity with Free Energy Perturbation","volume":"60","author":"Deflorian","year":"2020","journal-title":"J. Chem. Inf. Model."},{"issue":"2","key":"2026091409073413400_cit20","doi-asserted-by":"publisher","first-page":"293","DOI":"10.1021\/acsomega.6b00086","article-title":"Predicting Binding Affinities\nfor GPCR Ligands Using Free-Energy Perturbation","volume":"1","author":"Lenselink","year":"2016","journal-title":"ACS Omega."},{"issue":"12","key":"2026091409073413400_cit21","doi-asserted-by":"publisher","first-page":"6250","DOI":"10.1021\/cr3002609","article-title":"Modeling and\nSimulation of Ion Channels","volume":"112","author":"Maffeo","year":"2012","journal-title":"Chem. Rev."},{"key":"2026091409073413400_cit22","volume-title":"Predicting Secondary\nPharmacology with in Silico Structure-based Approaches","author":"Miller","year":"2025"},{"issue":"12","key":"2026091409073413400_cit23","doi-asserted-by":"publisher","first-page":"1792","DOI":"10.1016\/j.drudis.2017.08.004","article-title":"Structural\ncoverage of the proteome\nfor pharmaceutical applications","volume":"22","author":"Somody","year":"2017","journal-title":"Drug Discovery\nToday"},{"issue":"11","key":"2026091409073413400_cit24","doi-asserted-by":"publisher","first-page":"691","DOI":"10.1038\/nchembio.117","article-title":"Targeted\npolypharmacology:\ndiscovery of dual inhibitors of tyrosine and phosphoinositide kinases","volume":"4","author":"Apsel","year":"2008","journal-title":"Nat. Chem. Biol."},{"issue":"12","key":"2026091409073413400_cit25","doi-asserted-by":"publisher","first-page":"1489","DOI":"10.3390\/cancers12061489","article-title":"c-Src and EGFR Inhibition\nin Molecular Cancer Therapy:\nWhat Else Can We Improve?","volume":"12","author":"Belli","year":"2020","journal-title":"Cancers"},{"key":"2026091409073413400_cit26","doi-asserted-by":"crossref","unstructured":"Bowers, K.; Chow, E.; Xu, H.; et al.\n          Molecular dynamics---Scalable\nalgorithms for molecular dynamics simulations on commodity clusters, presented at:  Proceedings of the ACM\/IEEE Conference on\nSupercomputing (SC06)2006.","DOI":"10.1145\/1188455.1188544"},{"key":"2026091409073413400_cit27","unstructured":"Desmond Free Energy Perturbation\n(FEP) Package, Academic Release, D. E. Shaw\nResearch, 2023\u20134."},{"key":"2026091409073413400_cit28","doi-asserted-by":"publisher","first-page":"26","DOI":"10.1016\/j.phrs.2015.10.021","article-title":"Classification\nof small molecule protein kinase inhibitors based upon the structures\nof their drug-enzyme complexes","volume":"103","author":"Roskoski","year":"2016","journal-title":"Pharmacol. Res."},{"key":"2026091409073413400_cit29","doi-asserted-by":"publisher","first-page":"1625","DOI":"10.1021\/acs.accounts.7b00083","article-title":"Advancing\nDrug Discovery through Enhanced Free Energy Calculations","volume":"50","author":"Abel","year":"2017","journal-title":"Acc. Chem. Res."},{"issue":"1","key":"2026091409073413400_cit30","doi-asserted-by":"publisher","DOI":"10.1186\/s13321-020-0408-x","article-title":"Towards reproducible computational\ndrug discovery","volume":"12","author":"Schaduangrat","year":"2020","journal-title":"J. Cheminform."},{"issue":"3","key":"2026091409073413400_cit31","doi-asserted-by":"publisher","first-page":"638","DOI":"10.1021\/acs.jmedchem.7b00681","article-title":"Free Energy\nMethods in Drug Design:\nProspects of \u201cAlchemical Perturbation\u201d in Medicinal\nChemistry","volume":"61","author":"Williams-Noonan","year":"2018","journal-title":"J. Med. Chem."},{"issue":"6","key":"2026091409073413400_cit32","doi-asserted-by":"publisher","first-page":"2485","DOI":"10.1021\/acs.jmedchem.6b01881","article-title":"Prospective Evaluation\nof Free Energy Calculations for the Prioritization of Cathepsin L\nInhibitors","volume":"60","author":"Kuhn","year":"2017","journal-title":"J. Med. Chem."},{"issue":"10","key":"2026091409073413400_cit33","doi-asserted-by":"publisher","first-page":"6259","DOI":"10.1021\/acs.jctc.2c00752","article-title":"Chemical Space\nExploration with Active Learning and Alchemical Free Energies","volume":"18","author":"Khalak","year":"2022","journal-title":"J. Chem. Theory Comput."},{"key":"2026091409073413400_cit34","doi-asserted-by":"publisher","DOI":"10.1016\/j.ailsci.2022.100050","article-title":"Optimizing active learning\nfor free energy calculations","volume":"2","author":"Thompson","year":"2022","journal-title":"Artif. Intell.\nLife Sci."},{"issue":"9","key":"2026091409073413400_cit35","doi-asserted-by":"publisher","first-page":"3782","DOI":"10.1021\/acs.jcim.9b00367","article-title":"Reaction-Based Enumeration,\nActive Learning, and Free Energy Calculations To Rapidly Explore Synthetically\nTractable Chemical Space and Optimize Potency of Cyclin-Dependent\nKinase 2 Inhibitors","volume":"59","author":"Konze","year":"2019","journal-title":"J. Chem. Theory Comput."},{"issue":"11","key":"2026091409073413400_cit36","doi-asserted-by":"publisher","first-page":"2115","DOI":"10.3390\/molecules24112115","article-title":"Error Tolerance of Machine Learning\nAlgorithms across\nContemporary Biological Targets","volume":"24","author":"Kaiser","year":"2019","journal-title":"Molecules"},{"issue":"1","key":"2026091409073413400_cit37","doi-asserted-by":"publisher","first-page":"477","DOI":"10.1021\/acs.jctc.3c00839","article-title":"Using AlphaFold and\nExperimental Structures for the Prediction of the Structure and Binding\nAffinities of GPCR Complexes via Induced Fit Docking and Free Energy\nPerturbation","volume":"20","author":"Coskun","year":"2024","journal-title":"J. Chem. Theory Comput."},{"issue":"4","key":"2026091409073413400_cit38","doi-asserted-by":"publisher","first-page":"2630","DOI":"10.1021\/acs.jctc.1c00136","article-title":"Reliable and Accurate\nSolution to the Induced Fit Docking Problem for Protein\u2013Ligand\nBinding","volume":"17","author":"Miller","year":"2021","journal-title":"J. Chem. Theory Comput."},{"issue":"14","key":"2026091409073413400_cit39","doi-asserted-by":"publisher","first-page":"2895","DOI":"10.1021\/jm0300330","article-title":"Information Decay in Molecular Docking Screens against\nHolo, Apo, and Modeled Conformations of Enzymes","volume":"46","author":"McGovern","year":"2003","journal-title":"J. Med. Chem."},{"issue":"1","key":"2026091409073413400_cit40","doi-asserted-by":"publisher","DOI":"10.1002\/pro.4530","article-title":"Evaluation of AlphaFold2\nstructures as docking targets","volume":"32","author":"Holcomb","year":"2023","journal-title":"Protein Sci."},{"key":"2026091409073413400_cit41","volume-title":"Schro\u0308dinger\nRelease 2024\u20133, Maestro","year":"2024"},{"key":"2026091409073413400_cit42","unstructured":"Lim, N. M.; Wang, L.; Abel, R.; Mobley, D. L.\n          Sensitivity\nin Binding Free Energies\nDue to Protein Reorganization. (1549\u20139626 (Electronic))."},{"issue":"2","key":"2026091409073413400_cit43","doi-asserted-by":"publisher","first-page":"754","DOI":"10.1021\/acs.jcim.8b00826","article-title":"Assessing\nthe Predictive Power of Relative Binding\nFree Energy Calculations for Test Cases Involving Displacement of\nBinding Site Water Molecules","volume":"59","author":"Wahl","year":"2019","journal-title":"J. Chem. Inf.\nModel."},{"key":"2026091409073413400_cit44","unstructured":"D. E. Shaw Research.\n          . Desmond Users Guide.\nversion 3611\/08. New York, NY, USA, 2014."}],"container-title":["Journal of Chemical Information\nand Modeling"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/pubs.acs.org\/doi\/pdf\/10.1021\/acs.jcim.6c01024","content-type":"application\/pdf","content-version":"vor","intended-application":"unspecified"},{"URL":"https:\/\/pubs.acs.org\/jcisd8\/article-pdf\/66\/17\/10992\/66889037\/acs.jcim.6c01024.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/pubs.acs.org\/jcisd8\/article-pdf\/66\/17\/10992\/66889037\/acs.jcim.6c01024.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,14]],"date-time":"2026-09-14T13:30:11Z","timestamp":1789392611000},"score":1,"resource":{"primary":{"URL":"https:\/\/pubs.acs.org\/jcisd8\/article\/66\/17\/10992\/5277344\/To-ML-Predict-or-Not-to-ML-Predict-The-Impact-of"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,20]]},"references-count":44,"journal-issue":{"issue":"17","published-online":{"date-parts":[[2026,8,20]]},"published-print":{"date-parts":[[2026,9,14]]}},"URL":"https:\/\/doi.org\/10.1021\/acs.jcim.6c01024","relation":{},"ISSN":["1549-9596","1549-960X"],"issn-type":[{"value":"1549-9596","type":"print"},{"value":"1549-960X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,20]]}}}