{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,19]],"date-time":"2026-09-19T16:03:37Z","timestamp":1789833817608,"version":"4.0.1"},"reference-count":65,"publisher":"American Chemical Society (ACS)","issue":"17","license":[{"start":{"date-parts":[[2026,8,14]],"date-time":"2026-08-14T00:00:00Z","timestamp":1786665600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100004350","name":"Studienstiftung des Deutschen Volkes","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100004350","id-type":"DOI","asserted-by":"publisher"}]}],"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>Machine learning has emerged as a powerful approach for molecular property prediction and drug discovery. However, the black-box nature of many machine learning models limits their interpretability, trustworthiness, and adoption in interdisciplinary research settings. This is especially the case in molecular machine learning, where explanations of model predictions are used for informed decision-making and downstream tasks. Shapley values, originating from cooperative game theory, provide a principled framework for attributing model predictions to individual input features. However, existing Shapley value-based explanations for molecular machine learning often rely on sampling-based approximations or operate at the level of abstract features, which can reduce attribution stability and limit chemical interpretability and actionability. Here, we introduce a fragment-level Shapley value framework that enables the exact computation of feature contributions at the level of chemically meaningful fragments for molecular property predictions without relying on sampling or feature imputation. By decomposing molecules into fragments, the proposed approach yields actionable explanations that can be directly related to established chemical concepts. We apply the method post hoc to random forest and graph convolutional network models using common molecular representations, including extended-connectivity fingerprints and molecular graphs. The approach is evaluated across three representative property prediction tasks: aqueous solubility, mutagenicity, and antiviral potency. Fragment-level Shapley values reproduce well-established chemical trends, identify known toxicophores, and enable guided molecular optimization. In addition, the method provides insights into model learning characteristics and helps delineate the applicability domain, particularly in settings with limited and structurally biased data. Overall, this work demonstrates that adapting Shapley values to chemically meaningful fragments enables interpretable explanations for molecular machine learning models, supporting molecular optimization and model validation.<\/jats:p>","DOI":"10.1021\/acs.jcim.6c01425","type":"journal-article","created":{"date-parts":[[2026,8,15]],"date-time":"2026-08-15T01:01:12Z","timestamp":1786755672000},"page":"10661-10673","source":"Crossref","is-referenced-by-count":1,"title":["Chemically Interpretable\nExplanations for Molecular\nProperty Prediction via Fragment-Level Shapley Values"],"prefix":"10.1021","volume":"66","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-6690-5601","authenticated-orcid":true,"given":"Jannik P.","family":"Roth","sequence":"first","affiliation":[{"name":"University of Bonn , , ,","place":["Bonn, Germany, 53115"]},{"name":"University of Bonn , , ,","place":["Bonn, Germany, 53115"]},{"name":"University of Bonn , , ,","place":["Bonn, Germany, 53115"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"316","published-online":{"date-parts":[[2026,8,14]]},"reference":[{"key":"2026091409074751400_cit1","doi-asserted-by":"publisher","first-page":"547","DOI":"10.1038\/s41586-018-0337-2","article-title":"Machine learning\nfor molecular and materials science","volume":"559","author":"Butler","year":"2018","journal-title":"Nature"},{"key":"2026091409074751400_cit2","doi-asserted-by":"publisher","first-page":"463","DOI":"10.1038\/s41573-019-0024-5","article-title":"Applications\nof machine learning in drug discovery and development","volume":"18","author":"Vamathevan","year":"2019","journal-title":"Nat. Rev. Drug Discovery"},{"key":"2026091409074751400_cit3","doi-asserted-by":"publisher","first-page":"43","DOI":"10.1146\/annurev-biodatasci-122120-124216","article-title":"Machine learning in chemoinformatics\nand medicinal chemistry","volume":"5","author":"Rodri\u0301guez-Pe\u0301rez","year":"2022","journal-title":"Annu. Rev. Biomed.\nData Sci."},{"key":"2026091409074751400_cit4","doi-asserted-by":"publisher","first-page":"1538","DOI":"10.1016\/j.drudis.2018.05.010","article-title":"Machine learning\nin chemoinformatics and drug discovery","volume":"23","author":"Lo","year":"2018","journal-title":"Drug\nDiscovery Today"},{"key":"2026091409074751400_cit5","doi-asserted-by":"publisher","first-page":"1241","DOI":"10.1016\/j.drudis.2018.01.039","article-title":"The rise of deep learning in drug\ndiscovery","volume":"23","author":"Chen","year":"2018","journal-title":"Drug Discovery Today"},{"key":"2026091409074751400_cit6","doi-asserted-by":"publisher","DOI":"10.2144\/fsoa-2022-0010","article-title":"Artificial intelligence\nin interdisciplinary life science and drug discovery research","volume":"8","author":"Bajorath","year":"2022","journal-title":"Future Sci. OA"},{"key":"2026091409074751400_cit7","doi-asserted-by":"publisher","first-page":"967","DOI":"10.1016\/j.drudis.2021.11.023","article-title":"Enhancing\npreclinical drug discovery with artificial intelligence","volume":"27","author":"Vijayan","year":"2022","journal-title":"Drug Discovery Today"},{"key":"2026091409074751400_cit8","doi-asserted-by":"publisher","first-page":"100056","DOI":"10.1016\/j.ailsci.2022.100056","article-title":"Machine learning\nfor small molecule drug discovery\nin academia and industry","volume":"3","author":"Volkamer","year":"2023","journal-title":"Artif. Intell. Life\nSci."},{"key":"2026091409074751400_cit9","doi-asserted-by":"publisher","first-page":"5975","DOI":"10.1007\/s10462-022-10306-1","article-title":"Deep learning\nin drug discovery:\nan integrative review and future challenges","volume":"56","author":"Askr","year":"2023","journal-title":"Artificial Intelligence Review"},{"key":"2026091409074751400_cit10","doi-asserted-by":"publisher","first-page":"20","DOI":"10.1038\/538020a","article-title":"Can we open\nthe black box of AI?","volume":"538","author":"Castelvecchi","year":"2016","journal-title":"Nature"},{"key":"2026091409074751400_cit11","doi-asserted-by":"publisher","first-page":"81","DOI":"10.1038\/s43586-022-00172-0","article-title":"Why black\nbox machine learning should be avoided for high-stakes decisions,\nin brief","volume":"2","author":"Rudin","year":"2022","journal-title":"Nat. Rev. Methods Primers"},{"key":"2026091409074751400_cit12","doi-asserted-by":"publisher","DOI":"10.1126\/scirobotics.aay7120","article-title":"XAI\u2500Explainable artificial\nintelligence","volume":"4","author":"Gunning","year":"2019","journal-title":"Science Robotics"},{"key":"2026091409074751400_cit13","doi-asserted-by":"publisher","first-page":"688969","DOI":"10.3389\/fdata.2021.688969","article-title":"Principles and practice of explainable\nmachine learning","volume":"4","author":"Belle","year":"2021","journal-title":"Frontiers in Big Data"},{"key":"2026091409074751400_cit14","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3236009","article-title":"A survey of methods for explaining\nblack box models","volume":"51","author":"Guidotti","year":"2019","journal-title":"ACM Computing Surveys"},{"key":"2026091409074751400_cit15","doi-asserted-by":"publisher","DOI":"10.1002\/widm.1424","article-title":"Explainable\nartificial intelligence: an analytical review","volume":"11","author":"Angelov","year":"2021","journal-title":"WIREs Data Mining and Knowledge Discovery"},{"key":"2026091409074751400_cit16","doi-asserted-by":"publisher","first-page":"101805","DOI":"10.1016\/j.inffus.2023.101805","article-title":"Explainable artificial intelligence (XAI): what we\nknow and what is left to attain trustworthy artificial intelligence","volume":"99","author":"Ali","year":"2023","journal-title":"Information Fusion"},{"key":"2026091409074751400_cit17","doi-asserted-by":"publisher","first-page":"18","DOI":"10.3390\/e23010018","article-title":"Explainable AI: a review of machine\nlearning interpretability methods","volume":"23","author":"Linardatos","year":"2021","journal-title":"Entropy"},{"key":"2026091409074751400_cit18","article-title":"A unified approach\nto interpreting model predictions","volume-title":"Advances\nin Neural Information Processing Systems","author":"Lundberg","year":"2017"},{"key":"2026091409074751400_cit19","doi-asserted-by":"publisher","first-page":"56","DOI":"10.1038\/s42256-019-0138-9","article-title":"From local explanations to global\nunderstanding with explainable AI for trees","volume":"2","author":"Lundberg","year":"2020","journal-title":"Nature Machine Intelligence"},{"key":"2026091409074751400_cit20","article-title":"GNNExplainer:\ngenerating explanations\nfor graph neural networks","volume-title":"Advances in\nNeural Information Processing Systems","author":"Ying","year":"2019"},{"key":"2026091409074751400_cit21","first-page":"307","volume-title":"A value for n-person\ngames In Contributions to the Theory of Games","author":"Shapley","year":"1953"},{"key":"2026091409074751400_cit22","doi-asserted-by":"publisher","first-page":"590","DOI":"10.1038\/s42256-023-00657-x","article-title":"Algorithms\nto estimate\nShapley value feature attributions","volume":"5","author":"Chen","year":"2023","journal-title":"Nature Machine\nIntelligence"},{"key":"2026091409074751400_cit23","doi-asserted-by":"publisher","first-page":"17744","DOI":"10.1021\/acs.jmedchem.1c01789","article-title":"Explainable machine learning for property predictions\nin compound optimization: miniperspective","volume":"64","author":"Rodri\u0301guez-Pe\u0301rez","year":"2021","journal-title":"J.\nMed. Chem."},{"key":"2026091409074751400_cit24","doi-asserted-by":"publisher","first-page":"647","DOI":"10.1007\/s10115-013-0679-x","article-title":"Explaining\nprediction models and individual predictions\nwith feature contributions","volume":"41","author":"S\u030ctrumbelj","year":"2014","journal-title":"Knowledge and Information\nSystems"},{"key":"2026091409074751400_cit25","doi-asserted-by":"crossref","unstructured":"Datta, A.; Sen, S.; Zick, Y.\n          Algorithmic transparency\nvia quantitative\ninput influence: theory and experiments with learning systems. In  2016 IEEE Symposium on Security and Privacy (SP); IEEE, 2016; pp 598\u2013617.10.1109\/sp.2016.42.","DOI":"10.1109\/SP.2016.42"},{"key":"2026091409074751400_cit26","doi-asserted-by":"publisher","first-page":"105023","DOI":"10.1016\/j.isci.2022.105023","article-title":"Calculation of exact\nShapley values for support vector\nmachines with Tanimoto kernel enables model interpretation","volume":"25","author":"Feldmann","year":"2022","journal-title":"iScience"},{"key":"2026091409074751400_cit27","doi-asserted-by":"publisher","first-page":"19561","DOI":"10.1038\/s41598-023-46930-2","article-title":"Calculation\nof exact Shapley values\nfor explaining support vector machine models using the radial basis\nfunction kernel","volume":"13","author":"Mastropietro","year":"2023","journal-title":"Sci. Rep."},{"key":"2026091409074751400_cit28","doi-asserted-by":"publisher","first-page":"102110","DOI":"10.1016\/j.xcrp.2024.102110","article-title":"Machine learning models with distinct Shapley value\nexplanations decouple feature attribution and interpretation for chemical\ncompound predictions","volume":"5","author":"Roth","year":"2024","journal-title":"Cell Rep. Phys. Sci."},{"key":"2026091409074751400_cit29","unstructured":"Janzing, D.; Minorics, L.; Bloebaum, P.\n          Feature\nrelevance\nquantification in explainable AI: a causal problem. In  Proceedings of the Twenty Third International Conference\non Artificial Intelligence and Statistics; Chiappa, S., Calandra, R., Eds.; PMLR, 2020; pp 2907\u20132916."},{"key":"2026091409074751400_cit30","doi-asserted-by":"publisher","DOI":"10.1002\/cmdc.202300586","article-title":"Generation of molecular\ncounterfactuals for explainable\nmachine learning based on core-substituent recombination","volume":"19","author":"Lamens","year":"2024","journal-title":"ChemMedChem."},{"key":"2026091409074751400_cit31","doi-asserted-by":"publisher","first-page":"5601","DOI":"10.3390\/molecules28145601","article-title":"Explaining multiclass compound activity predictions\nusing counterfactuals and Shapley values","volume":"28","author":"Lamens","year":"2023","journal-title":"Molecules"},{"key":"2026091409074751400_cit32","doi-asserted-by":"publisher","first-page":"1547","DOI":"10.1039\/D4MD00128A","article-title":"Systematic generation and analysis\nof counterfactuals\nfor compound activity predictions using multi-task models","volume":"15","author":"Lamens","year":"2024","journal-title":"RSC Med. Chem."},{"key":"2026091409074751400_cit33","doi-asserted-by":"publisher","first-page":"143","DOI":"10.1186\/s13321-025-01100-6","article-title":"Contrastive explanations\nfor machine learning predictions in chemistry","volume":"17","author":"Lamens","year":"2025"},{"key":"2026091409074751400_cit34","doi-asserted-by":"publisher","first-page":"100230","DOI":"10.1016\/j.ejmcr.2024.100230","article-title":"MolAnchor method for explaining compound\npredictions\nbased on substructures","volume":"12","author":"Lamens","year":"2024","journal-title":"Eur. J. Med. Chem. Rep."},{"key":"2026091409074751400_cit35","doi-asserted-by":"publisher","first-page":"1357","DOI":"10.1021\/acs.jcim.4c02153","article-title":"Rationalizing predictions of isoform-selective\nphosphoinositide\n3-kinase inhibitors using MolAnchor analysis","volume":"65","author":"Lamens","year":"2025","journal-title":"J. Chem. Inf. Model."},{"key":"2026091409074751400_cit36","doi-asserted-by":"publisher","first-page":"1411","DOI":"10.1039\/D5SC08461J","article-title":"Explainable artificial intelligence for molecular design\nin pharmaceutical research","volume":"17","author":"Lamens","year":"2026","journal-title":"Chem. Sci."},{"key":"2026091409074751400_cit37","doi-asserted-by":"publisher","first-page":"2236","DOI":"10.1021\/acs.jcim.3c00465","article-title":"MolSHAP: interpreting\nquantitative structure\u2013activity relationships using Shapley\nvalues of R-groups","volume":"64","author":"Tian","year":"2024","journal-title":"J. Chem. Inf. Model."},{"key":"2026091409074751400_cit38","doi-asserted-by":"publisher","first-page":"2585","DOI":"10.1038\/s41467-023-38192-3","article-title":"Chemistry-intuitive explanation of graph neural networks for molecular\nproperty prediction with substructure masking","volume":"14","author":"Wu","year":"2023","journal-title":"Nat. Commun."},{"key":"2026091409074751400_cit39","doi-asserted-by":"publisher","first-page":"881","DOI":"10.1021\/acs.jcim.5c02225","article-title":"TradePool:\na novel interpretable\nframework for quantifying atomic attribution values in molecular property\nprediction","volume":"66","author":"Ni","year":"2026","journal-title":"J. Chem. Inf. Model."},{"key":"2026091409074751400_cit40","doi-asserted-by":"publisher","first-page":"7516","DOI":"10.1021\/acs.jcim.5c01003","article-title":"Explainable\nGraph Neural Networks in Chemistry: Combining Attribution and Uncertainty\nQuantification","volume":"65","author":"Komissarov","year":"2025","journal-title":"J. Chem. Inf. Model."},{"key":"2026091409074751400_cit41","doi-asserted-by":"publisher","first-page":"742","DOI":"10.1021\/ci100050t","article-title":"Extended-connectivity fingerprints","volume":"50","author":"Rogers","year":"2010","journal-title":"J. Chem.\nInf. Model."},{"key":"2026091409074751400_cit42","doi-asserted-by":"publisher","DOI":"10.1351\/goldbook.MT07069","volume-title":"Molecular Graph In Compendium of Chemical Terminology\n(the IUPAC\nGold Book)","author":"McNaught","year":"2014"},{"key":"2026091409074751400_cit43","doi-asserted-by":"publisher","first-page":"1503","DOI":"10.1002\/cmdc.200800178","article-title":"On the art of compiling\nand using \u201cdrug-like\u201d chemical fragment spaces","volume":"3","author":"Degen","year":"2008","journal-title":"ChemMedChem."},{"key":"2026091409074751400_cit44","doi-asserted-by":"publisher","DOI":"10.5281\/zenodo.17232453","volume-title":"rdkit\/rdkit:\n2025_09_1 (Q3 2025) Release, Version Release_2025_09_1","author":"Landrum","year":"2025"},{"key":"2026091409074751400_cit45","doi-asserted-by":"publisher","first-page":"511","DOI":"10.1021\/ci970429i","article-title":"RECAP \u2013 retrosynthetic\ncombinatorial analysis procedure: a powerful new technique for identifying\nprivileged molecular fragments with useful applications in combinatorial\nchemistry","volume":"38","author":"Lewell","year":"1998","journal-title":"J. Chem. Inf. Comput. Sci."},{"key":"2026091409074751400_cit46","doi-asserted-by":"publisher","first-page":"318","DOI":"10.1021\/ci6003652","article-title":"Chemical fragment spaces for de novo design","volume":"47","author":"Mauser","year":"2007","journal-title":"J. Chem. Inf. Model."},{"key":"2026091409074751400_cit47","unstructured":"Antiviral Drug Discorvery 2025 ASAP Discovery\nInitiative, https:\/\/polarishub.io\/competitions\/asap-discovery\/antiviral-drug-discovery-2025 (accessed Jan 26, 2026)."},{"key":"2026091409074751400_cit48","doi-asserted-by":"publisher","first-page":"1000","DOI":"10.1021\/ci034243x","article-title":"ESOL: estimating\naqueous solubility directly from molecular structure","volume":"44","author":"Delaney","year":"2004","journal-title":"J. Chem. Inf. Comput. Sci."},{"key":"2026091409074751400_cit49","doi-asserted-by":"publisher","first-page":"2077","DOI":"10.1021\/ci900161g","article-title":"Benchmark\ndata set for in silico prediction of Ames mutagenicity","volume":"49","author":"Hansen","year":"2009","journal-title":"J. Chem. Inf. Model."},{"key":"2026091409074751400_cit50","doi-asserted-by":"publisher","first-page":"29","DOI":"10.1016\/S0027-5107(00)00064-6","article-title":"The Ames Salmonella\/microsome\nmutagenicity assay","volume":"455","author":"Mortelmans","year":"2000","journal-title":"Mutat. Res., Fundam. Mol.\nMech. Mutagen."},{"key":"2026091409074751400_cit51","doi-asserted-by":"publisher","first-page":"5","DOI":"10.1023\/A:1010933404324","article-title":"Random forests","volume":"45","author":"Breiman","year":"2001","journal-title":"Machine Learning"},{"key":"2026091409074751400_cit52","doi-asserted-by":"publisher","first-page":"2825","DOI":"10.48550\/arXiv.1201.0490","article-title":"Scikit-learn: machine\nlearning in Python","volume":"12","author":"Pedregosa","year":"2011","journal-title":"Journal\nof Machine Learning Research"},{"key":"2026091409074751400_cit53","unstructured":"Fey, M.; Lenssen, J. E.\n          Fast graph\nrepresentation learning with PyTorch Geometric. In  ICLR Workshop on Representation Learning on Graphs and Manifolds, 2019."},{"key":"2026091409074751400_cit54","unstructured":"Fey, M.; Sunil, J.; Nitta, A.; Puri, R.; Shah, M.; Stojanovic\u030c, B.; Bendias, R.; Alexandria, B.; Kocijan, V.; Zhang, Z.; He, X.; Lenssen, J. E.; Leskovec, J.\n          PyG\n2.0: scalable learning on real world graphs. In  Temporal Graph Learning Workshop; KDD, 2025."},{"key":"2026091409074751400_cit55","doi-asserted-by":"crossref","unstructured":"Ansel, J.; et al.; PyTorch 2: faster machine learning through dynamic python bytecode\ntransformation and graph compilation. In  Proceedings\nof the 29th ACM International Conference on Architectural Support\nfor Programming Languages and Operating Systems; ACM, 2024; Vol. 2,\npp 929\u2013947.10.1145\/3620665.3640366.","DOI":"10.1145\/3620665.3640366"},{"key":"2026091409074751400_cit56","unstructured":"Kingma, D. P.; Ba, J.\n          Adam: A method for stochastic\noptimization. In  3rd International Conference\non Learning Representations, ICLR 2015, San Diego, CA, USA, May 7\u20139,\n2015, Conference Track Proceedings; Bengio, Y., LeCun, Y., Eds., 2015."},{"key":"2026091409074751400_cit57","doi-asserted-by":"publisher","first-page":"579","DOI":"10.2307\/2340126","article-title":"On the\nmethods of measuring association between two attributes","volume":"75","author":"Yule","year":"1912","journal-title":"Journal of the Royal Statistical Society"},{"key":"2026091409074751400_cit58","unstructured":"Antiviral Drug Discovery\n2025-Competition Results ASAP Discovery\nInitiative, https:\/\/polarishub.io\/competitions\/asap-discovery\/antiviral-drug-discovery-2025#competiton-results (accessed June 02, 2026)."},{"key":"2026091409074751400_cit59","doi-asserted-by":"publisher","first-page":"2962","DOI":"10.1021\/mp500103r","article-title":"Is Experimental\nData Quality the Limiting Factor in\nPredicting the Aqueous Solubility of Druglike Molecules?","volume":"11","author":"Palmer","year":"2014","journal-title":"Mol. Pharmaceutics"},{"key":"2026091409074751400_cit60","doi-asserted-by":"publisher","DOI":"10.1201\/EBK1439802458","volume-title":"Handbook of Aqueous Solubility Data","author":"Yalkowsky","year":"2016"},{"key":"2026091409074751400_cit61","doi-asserted-by":"publisher","first-page":"2887","DOI":"10.1021\/jm9602928","article-title":"The Properties of Known Drugs. 1.\nMolecular Frameworks","volume":"39","author":"Bemis","year":"1996","journal-title":"J. Med. Chem."},{"key":"2026091409074751400_cit62","doi-asserted-by":"publisher","first-page":"312","DOI":"10.1021\/jm040835a","article-title":"Derivation and validation of toxicophores\nfor mutagenicity prediction","volume":"48","author":"Kazius","year":"2005","journal-title":"J. Med. Chem."},{"key":"2026091409074751400_cit63","doi-asserted-by":"publisher","first-page":"119","DOI":"10.1186\/s13321-023-00787-9","article-title":"SIMPD: an algorithm for generating\nsimulated time splits for validating machine learning approaches","volume":"15","author":"Landrum","year":"2023","journal-title":"J. Cheminf."},{"key":"2026091409074751400_cit64","doi-asserted-by":"publisher","first-page":"5996","DOI":"10.21105\/joss.05996","article-title":"Machine\nlearning validation via rational dataset sampling with astartes","volume":"8","author":"Burns","year":"2023","journal-title":"Journal of Open Source Software"},{"key":"2026091409074751400_cit65","doi-asserted-by":"crossref","unstructured":"Tricarico, G. A.; Hofmans, J.; Lenselink, E. B.; Lo\u0301pez-Ramos, M.; Dre\u0301anic, M.-P.; Stouten, P. F. W.\n          Construction of balanced, chemically dissimilar training. In  Validation and Test Sets for Machine Learning on Molecular\nDatasets, 2022.10.26434\/chemrxiv-2022-m8l33-v2.","DOI":"10.26434\/chemrxiv-2022-m8l33-v2"}],"container-title":["Journal of Chemical Information\nand Modeling"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/pubs.acs.org\/doi\/pdf\/10.1021\/acs.jcim.6c01425","content-type":"application\/pdf","content-version":"vor","intended-application":"unspecified"},{"URL":"https:\/\/pubs.acs.org\/jcisd8\/article-pdf\/66\/17\/10661\/66580318\/acs.jcim.6c01425.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/pubs.acs.org\/jcisd8\/article-pdf\/66\/17\/10661\/66580318\/acs.jcim.6c01425.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,14]],"date-time":"2026-09-14T13:30:25Z","timestamp":1789392625000},"score":1,"resource":{"primary":{"URL":"https:\/\/pubs.acs.org\/jcisd8\/article\/66\/17\/10661\/5257678\/Chemically-Interpretable-Explanations-for"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,14]]},"references-count":65,"journal-issue":{"issue":"17","published-online":{"date-parts":[[2026,8,14]]},"published-print":{"date-parts":[[2026,9,14]]}},"URL":"https:\/\/doi.org\/10.1021\/acs.jcim.6c01425","relation":{},"ISSN":["1549-9596","1549-960X"],"issn-type":[{"value":"1549-9596","type":"print"},{"value":"1549-960X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,14]]}}}