{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,14]],"date-time":"2026-09-14T14:19:13Z","timestamp":1789395553396,"version":"build-2803163510"},"reference-count":57,"publisher":"American Chemical Society (ACS)","issue":"17","license":[{"start":{"date-parts":[[2026,8,24]],"date-time":"2026-08-24T00:00:00Z","timestamp":1787529600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc-nd\/4.0\/"}],"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>Small-molecule drug discovery frequently operates in regimes where slight structural changes have large consequences. These are situations in which current artificial intelligence (AI) methods, trained on mass data, may perform poorly. For a substantial fraction of drug-discovery projects, limited biological understanding and sparse data further constrain progress and define regimes where AI approaches remain difficult to apply. Despite these limitations, the global investment in AI continues to grow rapidly. Using two case studies characterized by structural and biochemical data, this article examines how the close integration of computational and medicinal chemistry coupled with a deep understanding of chemical shape and protein interactions enabled targeted computational molecular design to produce disproportionate gains in efficacy and selectivity. These examples highlight a challenge for current AI methods: sensitivity to subtle, low-data perturbations. For AI to achieve a transformative impact in drug discovery, it must move beyond pattern recognition to understand, or explicitly simulate, the mechanistic \u201cwhy\u201d linking subtle structural changes to biological outcomes. Pending such advances, clear opportunities for AI to productively complement human efforts, rather than be used as a stand-alone solution, are delineated.<\/jats:p>","DOI":"10.1021\/acs.jcim.6c01513","type":"journal-article","created":{"date-parts":[[2026,9,14]],"date-time":"2026-09-14T13:06:41Z","timestamp":1789391201000},"page":"10450-10456","source":"Crossref","is-referenced-by-count":0,"title":["Is AI Capable of\nReal-World Drug Discovery?"],"prefix":"10.1021","volume":"66","author":[{"given":"Georgia","family":"McGaughey","sequence":"first","affiliation":[{"name":"Trilligant, LLC , , ,","place":["Santa Fe, New Mexico, United States, 87506"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"316","published-online":{"date-parts":[[2026,8,24]]},"reference":[{"key":"2026091409063723500_cit1","volume-title":"The Power\nof Myth","author":"Joseph Cambell","year":"1988"},{"issue":"8","key":"2026091409063723500_cit2","doi-asserted-by":"publisher","first-page":"2602","DOI":"10.1038\/s41591-025-03743-2","article-title":"A generative\nAI-discovered TNIK inhibitor for idiopathic pulmonary fibrosis: a\nrandomized phase 2a trial","volume":"31","author":"Xu","year":"2025","journal-title":"Nat. Med."},{"issue":"1","key":"2026091409063723500_cit3","doi-asserted-by":"publisher","first-page":"63","DOI":"10.1038\/s41587-024-02143-0","article-title":"A small-molecule TNIK inhibitor targets\nfibrosis in preclinical and clinical models","volume":"43","author":"Ren","year":"2025","journal-title":"Nat. Biotechnol."},{"key":"2026091409063723500_cit4","unstructured":"First AI-designed\ndrug, Rentosertib, officially named by USAN. Drug\nTarget Review. March 14, 2025."},{"key":"2026091409063723500_cit5","doi-asserted-by":"publisher","first-page":"147","DOI":"10.1089\/genedge.5.1.31","article-title":"BMS Collaboration\nPaying off for Exscientia","volume":"5","author":"Philippidis","year":"2023","journal-title":"GEN Edge"},{"key":"2026091409063723500_cit6","doi-asserted-by":"publisher","first-page":"1120","DOI":"10.1038\/s42256-024-00911-w","article-title":"A call for an industry-led initiative\nto critically assess machine learning for real-world drug discovery","volume":"6","author":"Wognum","year":"2024","journal-title":"Nat. Mach. Intell."},{"issue":"8","key":"2026091409063723500_cit7","doi-asserted-by":"publisher","first-page":"1169","DOI":"10.1021\/acsmedchemlett.4c00290","article-title":"Machine Learning\nADME Models in Practice: Four Guidelines from a Successful Lead Optimization\nCase Study","volume":"15","author":"Rich","year":"2024","journal-title":"ACS Med. Chem. Lett."},{"issue":"8","key":"2026091409063723500_cit8","doi-asserted-by":"publisher","first-page":"1269","DOI":"10.1021\/acsmedchemlett.4c00150","article-title":"AI Based\nDiscovery of a New AKR1C3 Inhibitor for Anticancer Applications","volume":"15","author":"Pippione","year":"2024","journal-title":"ACS Med. Chem. Lett."},{"key":"2026091409063723500_cit9","doi-asserted-by":"crossref","DOI":"10.1007\/978-981-16-5180-9_1","article-title":"History and Present\nScenario of Computers in Pharmaceutical Research and Development","volume-title":"Computer Aided Pharmaceutics and Drug Delivery","author":"Saharan","year":"2022"},{"issue":"12","key":"2026091409063723500_cit10","doi-asserted-by":"publisher","first-page":"4977","DOI":"10.1021\/jm4004285","article-title":"QSAR Modeling:\nWhere have you been? Where are you going to?","volume":"57","author":"Cherkasov","year":"2014","journal-title":"J. Med. Chem."},{"key":"2026091409063723500_cit11","doi-asserted-by":"publisher","first-page":"305","DOI":"10.1021\/jm00002a012","article-title":"A Priori Prediction\nof Activity for HIV-1 Protease Inhibitors Employing Energy Minimization\nin the Active Site","volume":"38","author":"Holloway","year":"1995","journal-title":"J. Med. Chem."},{"key":"2026091409063723500_cit12","doi-asserted-by":"publisher","first-page":"3544","DOI":"10.1021\/acs.jcim.5c00245","article-title":"Critical Assessment of RNA and DNA Structure Predictions\nvia Artificial Intelligence: The Imitation Game","volume":"65","author":"Bergonzo","year":"2025","journal-title":"J. Chem. Inf. Model."},{"key":"2026091409063723500_cit13","doi-asserted-by":"publisher","first-page":"637","DOI":"10.1038\/nsb0895-637","article-title":"The structural basis\nof aspirin activity\ninferred from the crystal structure of inactivated prostaglandin H2 synthase","volume":"2","author":"Loll","year":"1995","journal-title":"Nat. Struct. Mol. Biol."},{"issue":"2","key":"2026091409063723500_cit14","doi-asserted-by":"publisher","first-page":"310","DOI":"10.1074\/mcp.O116.065219","article-title":"A Proteomic Approach\nto Analyze the Aspirin-mediated Lysine Acetylome","volume":"16","author":"Tatham","year":"2017","journal-title":"Mol. Cell. Proteomics"},{"issue":"11","key":"2026091409063723500_cit15","doi-asserted-by":"publisher","first-page":"2997","DOI":"10.1016\/j.bmcl.2009.04.026","article-title":"Conformational analysis\nof N.N-disubstituted-1,4-diazepane orexin receptor antagonists and\nimplications for receptor binding","volume":"19","author":"Cox","year":"2009","journal-title":"Bioorg. Med.\nChem. Lett."},{"key":"2026091409063723500_cit16","doi-asserted-by":"publisher","first-page":"72","DOI":"10.1111\/fcp.12723","article-title":"Orexin receptors: Targets and applications","volume":"36","author":"Subramanian","year":"2022","journal-title":"Fundam. Clin. Pharmacol."},{"issue":"1","key":"2026091409063723500_cit17","doi-asserted-by":"publisher","first-page":"5","DOI":"10.1007\/s10822-014-9710-x","article-title":"Shaping\nsuvorexant: application of experimental and theoretical methods for\ndriving synthetic design","volume":"28","author":"McGaughey","year":"2014","journal-title":"J. Comput.-Aided Mol.\nDes."},{"key":"2026091409063723500_cit18","first-page":"192","article-title":"Structures\nof the human OX1 orexin receptor bound to selective and dual antagonists","volume":"12","author":"Yin","year":"2016","journal-title":"Nat. Struct. Mol. Biol."},{"key":"2026091409063723500_cit19","doi-asserted-by":"publisher","first-page":"1528","DOI":"10.1021\/acs.jmedchem.9b01787","article-title":"Crystal structure of the Orexin-1 receptor in complex\nwith filorexant","volume":"63","author":"Rappas","year":"2020","journal-title":"J. Med. Chem."},{"key":"2026091409063723500_cit20","doi-asserted-by":"crossref","first-page":"1582","DOI":"10.1016\/j.str.2022.11.001","article-title":"Crystal structure\nof the human OX2R bound to the insomnia\ndrug lemborexant","volume":"30","author":"Asada","year":"2022","journal-title":"Structure"},{"key":"2026091409063723500_cit21","doi-asserted-by":"publisher","first-page":"5061","DOI":"10.1021\/jm100112j","article-title":"A Medicinal Chemist\u2019s\nGuide\nto Molecular Interactions","volume":"53","author":"Bissantz","year":"2010","journal-title":"J. Med. Chem."},{"key":"2026091409063723500_cit22","doi-asserted-by":"publisher","first-page":"4489","DOI":"10.1021\/jm3003697","article-title":"Methyl\neffects on Protein-Ligand Binding","volume":"55","author":"Leung","year":"2012","journal-title":"J. Med. Chem."},{"key":"2026091409063723500_cit23","doi-asserted-by":"publisher","first-page":"1157","DOI":"10.3390\/ph16081157","article-title":"The magic methyl and its tricks in\ndrug discovery and development","volume":"16","author":"Pinheiro","year":"2023","journal-title":"Pharmaceuticals"},{"key":"2026091409063723500_cit24","doi-asserted-by":"publisher","first-page":"415","DOI":"10.1002\/cmdc.201200025","article-title":"Discovery of [(2R,5R)-5-{[(5-fluoropyridin-2-yl)oxy]methyl}-2-methylpiperidin-1-yl][5-methyl-2-(pyrimidin-2-yl)phenyl]methanone\n(MK-6096): a dual orexin receptor antagonist with potent sleep-promoting\nproperties","volume":"7","author":"Coleman","year":"2012","journal-title":"ChemMedChem"},{"key":"2026091409063723500_cit25","doi-asserted-by":"publisher","DOI":"10.1016\/j.bmc.2021.116308","article-title":"De novo design\nwith deep generative models based on 3D similarity scoring","volume":"46","author":"Papadopoulos","year":"2021","journal-title":"Bioorg. Med. Chem."},{"key":"2026091409063723500_cit26","doi-asserted-by":"publisher","first-page":"3388","DOI":"10.1021\/acs.jcim.5c00018","article-title":"Understanding conformation\nimportance in data-driven property prediction\nmodels","volume":"65","author":"Hamakawa","year":"2025","journal-title":"J. Chem. Inf. Model."},{"issue":"5","key":"2026091409063723500_cit27","doi-asserted-by":"publisher","first-page":"370","DOI":"10.1111\/j.1747-0285.2008.00717.x","article-title":"Comparative QSAR\nstudies of CYP1A2 inhibitor flavonoids\nusing 2D and 3D descriptors","volume":"72","author":"Roy","year":"2008","journal-title":"Chem. Biol. Drug\nDes."},{"issue":"2","key":"2026091409063723500_cit28","doi-asserted-by":"publisher","first-page":"1157","DOI":"10.1021\/acs.jmedchem.2c01521","article-title":"Discovery\nof MK-8189, a Highly Potent and Selective PDE10A Inhibitor\nfor the Treatment of Schizophrenia","volume":"66","author":"Layton","year":"2023","journal-title":"J. Med.\nChem."},{"key":"2026091409063723500_cit29","unstructured":"https:\/\/clinicaltrials.gov\/study\/NCT04624243?intr=MK-8189 (accessed\nJan 8, 2026)."},{"key":"2026091409063723500_cit30","volume-title":"The Book of Why: The\nNew Science of Cause and Effect","author":"Pearl","year":"2018"},{"key":"2026091409063723500_cit31","unstructured":"Scho\u0308lkopf, B.\n          \n          Causality\nfor Machine Learning, 2019. arXiv:1911.10500.\narXiv.org e-Print archive. https:\/\/arxiv.org\/abs\/1911.10500."},{"key":"2026091409063723500_cit32","doi-asserted-by":"publisher","DOI":"10.1016\/j.drudis.2023.103737","article-title":"Causal inference in drug discovery\nand development","volume":"28","author":"Michoel","year":"2023","journal-title":"Drug Discovery Today"},{"key":"2026091409063723500_cit33","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":"2026091409063723500_cit34","doi-asserted-by":"publisher","first-page":"1514","DOI":"10.1038\/s41592-024-02272-z","article-title":"OpenFold: Retraining\nAlphaFold2 yields new insights into its learning mechanisms and capacity\nfor generalization","volume":"21","author":"Ahdritz","year":"2024","journal-title":"Nat. Methods"},{"key":"2026091409063723500_cit35","doi-asserted-by":"publisher","DOI":"10.1101\/2025.06.14.659707","article-title":"Boltz-2: Towards\nAccurate and Efficient Binding Affinity Prediction","author":"Passaro","year":"2025","journal-title":"bioRxiv"},{"issue":"18","key":"2026091409063723500_cit36","doi-asserted-by":"publisher","first-page":"4331","DOI":"10.3390\/ijms20184331","article-title":"Molecular docking: shifting paradigms in drug discovery","volume":"20","author":"Pinzi","year":"2019","journal-title":"Int. J. Mol. Sci."},{"issue":"6","key":"2026091409063723500_cit37","doi-asserted-by":"publisher","first-page":"171","DOI":"10.1007\/s00894-019-4032-5","article-title":"Receptor-based pharmacophore\nmodeling, virtual screening, and molecular docking studies for the\ndiscovery of novel GSK-3\u03b2 inhibitors","volume":"25","author":"El Kerdawy","year":"2019","journal-title":"J. Mol. Model."},{"issue":"6","key":"2026091409063723500_cit38","doi-asserted-by":"publisher","first-page":"3265","DOI":"10.1021\/acs.jcim.0c00171","article-title":"In silico screening based discovery\nof novel inhibitors of human\ncyclic GMP\u2013AMP synthase: a cross-validation study of molecular\ndocking and experimental testing","volume":"60","author":"Zhao","year":"2020","journal-title":"J. Chem. Inf.\nModel."},{"key":"2026091409063723500_cit39","doi-asserted-by":"publisher","first-page":"224","DOI":"10.1038\/s41586-019-0917-9","article-title":"Ultra-large library docking for discovering new chemotypes","author":"Lyu","year":"2019","journal-title":"Nature"},{"key":"2026091409063723500_cit40","doi-asserted-by":"publisher","DOI":"10.64898\/2026.06.29.735309","article-title":"Identifying\nand Addressing Systematic Data Leakage in Protein-Ligand Affinity\nBenchmarks","author":"Mattsson","year":"2026","journal-title":"bioRxiv"},{"key":"2026091409063723500_cit41","doi-asserted-by":"publisher","DOI":"10.1186\/s13321-020-00460-5","article-title":"Molecular\nrepresentations\nin AI-driven drug discovery: a review and practical guide","volume":"12","author":"David","year":"2020","journal-title":"J. Cheminf."},{"key":"2026091409063723500_cit42","doi-asserted-by":"crossref","unstructured":"Vignac, C.; et al. In  MiDi:\nMixed Graph and 3D Denoising Diffusion for Molecule Generation, Proceedings of the 40th International Conference on Machine Learning\n(ICML), 2023.","DOI":"10.1007\/978-3-031-43415-0_33"},{"key":"2026091409063723500_cit43","unstructured":"Le, T. T.; et al.\n          EQGAT-diff: Equivariant Graph Attention Diffusion\nfor 3D Molecule\nGeneration. 2023, arXiv:2309.17296. arXiv.org\ne-Print archive. https:\/\/arxiv.org\/abs\/2309.17296."},{"key":"2026091409063723500_cit44","unstructured":"Dunn, I.; Koes, D. R. In FlowMol: Exploring\nFlow Matching for 3D De Novo Molecule Generation,\nMachine Learning for Structural Biology (MLSB), NeurIPS Workshop, 2024."},{"key":"2026091409063723500_cit45","doi-asserted-by":"publisher","first-page":"2052","DOI":"10.1039\/D5DD00363F","article-title":"FlowMol3: Flow Matching\nfor 3D De Novo Small-Molecule\nGeneration","volume":"5","author":"Dunn","year":"2026","journal-title":"Digital Discovery"},{"key":"2026091409063723500_cit46","unstructured":"Irwin, R.; Tibo, A.; Janet, J. P.; Olsson, S. In  SemlaFlow: Efficient\n3D Molecular\nGeneration with Latent Attention and Equivariant Flow Matching, Proceedings of the 28th International Conference on Artificial\nIntelligence and Statistics (AISTATS), PMLR 258, 2025; pp 3772\u20133780."},{"key":"2026091409063723500_cit47","doi-asserted-by":"publisher","first-page":"565","DOI":"10.1038\/s43588-026-00998-8","article-title":"FLOWR: Flow Matching for Structure-Aware De Novo, Interaction-\nand Fragment-Based Ligand Generation","volume":"6","author":"Cremer","year":"2026","journal-title":"Nat. Comput.\nSci."},{"key":"2026091409063723500_cit48","doi-asserted-by":"publisher","first-page":"14954","DOI":"10.1039\/D4SC03523B","article-title":"PILOT: Equivariant Diffusion\nfor Pocket-Conditioned de Novo Ligand Generation with Multi-Objective\nGuidance via Importance Sampling","volume":"15","author":"Cremer","year":"2024","journal-title":"Chem. Sci."},{"key":"2026091409063723500_cit49","unstructured":"Buttenschoen, M.; Ziv, Y.; Morris, G. M.; Deane, C. M.\n          An Evaluation\nof Unconditional 3D Molecular Generation Methods. 2025."},{"key":"2026091409063723500_cit50","doi-asserted-by":"publisher","first-page":"435","DOI":"10.1038\/s41563-019-0338-z","article-title":"Exploiting machine learning\nfor end-to-end drug discovery and development","volume":"18","author":"Ekins","year":"2019","journal-title":"Nat. Mater."},{"key":"2026091409063723500_cit51","doi-asserted-by":"publisher","first-page":"22858","DOI":"10.1002\/anie.201909987","article-title":"Autonomous discovery in the chemical\nsciences Part I: Progress","volume":"59","author":"Coley","year":"2020","journal-title":"Angew. Chem., Int.\nEd."},{"key":"2026091409063723500_cit52","doi-asserted-by":"publisher","first-page":"23414","DOI":"10.1002\/anie.201909989","article-title":"Autonomous discovery in the chemical\nsciences Part II: Outlook","volume":"59","author":"Coley","year":"2020","journal-title":"Angew. Chem., Int.\nEd."},{"key":"2026091409063723500_cit53","doi-asserted-by":"publisher","first-page":"203","DOI":"10.1038\/s41586-018-0579-z","article-title":"The UK Biobank resource\nwith deep phenotyping and genomic data","volume":"562","author":"Bycroft","year":"2018","journal-title":"Nature"},{"key":"2026091409063723500_cit54","doi-asserted-by":"publisher","DOI":"10.1126\/scitranslmed.adf4428","article-title":"Prospective study design\nand data analysis in UK Biobank","volume":"16","author":"Allen","year":"2024","journal-title":"Sci. Transl.\nMed."},{"key":"2026091409063723500_cit55","unstructured":"https:\/\/globalpatentsearch.ai\/10-leading-companies-using-ai-for-drug-discovery-2025\/ (accessed Jan 8, 2026)."},{"key":"2026091409063723500_cit56","unstructured":"https:\/\/wewillcure.com\/insights\/company-profiles\/ai-and-machine-learning\/innovative-companies-using-ai-for-drug-discovery (accessed Jan 8, 2026)."},{"key":"2026091409063723500_cit57","doi-asserted-by":"publisher","DOI":"10.1038\/s41573-026-01496-2","article-title":"Artificial\nintelligence in drug discovery\u2500what it is, where we stand and\nthe path forward","author":"Bender","year":"2026","journal-title":"Nat. Rev. Drug Discovery"}],"container-title":["Journal of Chemical Information\nand Modeling"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/pubs.acs.org\/doi\/pdf\/10.1021\/acs.jcim.6c01513","content-type":"application\/pdf","content-version":"vor","intended-application":"unspecified"},{"URL":"https:\/\/pubs.acs.org\/jcisd8\/article-pdf\/66\/17\/10450\/67365364\/acs.jcim.6c01513.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/pubs.acs.org\/jcisd8\/article-pdf\/66\/17\/10450\/67365364\/acs.jcim.6c01513.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,14]],"date-time":"2026-09-14T13:29:14Z","timestamp":1789392554000},"score":1,"resource":{"primary":{"URL":"https:\/\/pubs.acs.org\/jcisd8\/article\/66\/17\/10450\/5325489\/Is-AI-Capable-of-Real-World-Drug-Discovery"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,24]]},"references-count":57,"journal-issue":{"issue":"17","published-online":{"date-parts":[[2026,8,24]]},"published-print":{"date-parts":[[2026,9,14]]}},"URL":"https:\/\/doi.org\/10.1021\/acs.jcim.6c01513","relation":{},"ISSN":["1549-9596","1549-960X"],"issn-type":[{"value":"1549-9596","type":"print"},{"value":"1549-960X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,24]]}}}