{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,23]],"date-time":"2026-10-23T20:57:38Z","timestamp":1792789058423,"version":"4.5.0"},"reference-count":63,"publisher":"American Chemical Society (ACS)","issue":"39","license":[{"start":{"date-parts":[[2026,9,20]],"date-time":"2026-09-20T00:00:00Z","timestamp":1789862400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc-nd\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100001871","name":"Funda??o para a Ci?ncia e a Tecnologia","doi-asserted-by":"publisher","award":["UID\/00686\/2025"],"award-info":[{"award-number":["UID\/00686\/2025"]}],"id":[{"id":"10.13039\/501100001871","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001871","name":"Funda??o para a Ci?ncia e a Tecnologia","doi-asserted-by":"publisher","award":["UID\/50006\/2025"],"award-info":[{"award-number":["UID\/50006\/2025"]}],"id":[{"id":"10.13039\/501100001871","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,10,6]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Phosphodiesterase type 5A (PDE5A) inhibitors represent an important class of bioactive molecules with therapeutic relevance across cardiovascular, neurological, and oncological disorders. In this work, we develop an interpretable Quantitative Structure\u2013Activity Relationship (QSAR) framework that integrates classical machine learning (ML) approaches with a hybrid deep learning architecture (Transformer\u2013Convolutional Neural Network, Transformer\u2013CNN) to model enzyme inhibition using a curated and diverse data set from ChEMBL. Multiple structural descriptors and fingerprint-based representations were evaluated using established ML algorithms, with fingerprint-driven models, particularly those employing FCFP4 features, achieving the highest predictive performance with Matthews Correlation Coefficients up to 0.783. Comparative analysis revealed that optimized classical ML models slightly outperform the Transformer\u2013CNN approach for this data set. Model interpretability was explored through Shapley Additive Explanations (SHAP) analysis, similarity maps, and atom-wise relevance scores, providing mechanistic insights and highlighting the importance of molecular symmetry and topological features in determining inhibitory potency. To support practical deployment, the final models were implemented in a freely accessible web application built with Flask and Streamlit for rapid prediction of novel compounds\u2019 inhibitory potential. Overall, this work provides a transparent and reproducible QSAR workflow, clarifies the relative strengths of classical and deep learning approaches, and offers a practical tool to assist in the design of next-generation PDE5A inhibitors.<\/jats:p>","DOI":"10.1021\/acsomega.6c03768","type":"journal-article","created":{"date-parts":[[2026,9,20]],"date-time":"2026-09-20T15:02:21Z","timestamp":1789916541000},"page":"58896-58907","source":"Crossref","is-referenced-by-count":1,"title":["Interpretable Machine\nLearning Frameworks for QSAR\nModeling: A Comparative Study of Classical Algorithms and Transformer-CNN\nArchitectures"],"prefix":"10.1021","volume":"11","author":[{"given":"Fatima Ezzahra","family":"Belkhanchi","sequence":"first","affiliation":[{"name":"Sultan Moulay Slimane University , , ,","place":["Beni Mellal, Morocco, BP 523"]},{"name":"University of Porto , , ,","place":["Porto, Portugal, 4169-007"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mohamed","family":"Mbarki","sequence":"additional","affiliation":[{"name":"Sultan Moulay Slimane University , , ,","place":["Beni Mellal, Morocco, BP 523"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Samir","family":"Chtita","sequence":"additional","affiliation":[{"name":"Hassan II University of Casablanca , , ,","place":["Casablanca, Morocco, B.P 7955"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Amit Kumar","family":"Halder","sequence":"additional","affiliation":[{"name":"University of Porto , , ,","place":["Porto, Portugal, 4169-007"]},{"name":"Dr. B. C. Roy College of Pharmacy and Allied Health Sciences, Dr. Meghnad Saha Sarani, Bidhannagar , , ,","place":["Durgapur, West Bengal, India, 713206"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3375-8670","authenticated-orcid":true,"given":"M. Nata\u0301lia D. S.","family":"Cordeiro","sequence":"additional","affiliation":[{"name":"University of Porto , , ,","place":["Porto, Portugal, 4169-007"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"316","published-online":{"date-parts":[[2026,9,20]]},"reference":[{"key":"2026100610484031900_cit1","doi-asserted-by":"publisher","DOI":"10.1016\/j.biopha.2020.111128","article-title":"Phosphodiesterase 5 (PDE5): Structure-function\nregulation and therapeutic applications of inhibitors","volume":"134","author":"Ahmed","year":"2021","journal-title":"Biomed. Pharmacother."},{"key":"2026100610484031900_cit2","doi-asserted-by":"publisher","DOI":"10.3389\/fchem.2024.1400886","article-title":"Mechanism\ninvestigation of highly selective inhibitors toward phosphodiesterase\n5 and 6 via the in vitro calculation and simulation","volume":"12","author":"Qu","year":"2024","journal-title":"Front. Chem."},{"key":"2026100610484031900_cit3","doi-asserted-by":"publisher","first-page":"674","DOI":"10.1038\/nrd893","article-title":"Phosphodiesterase\n5 inhibitors: current status and potential applications","volume":"1","author":"Rotella","year":"2002","journal-title":"Nat. Rev. Drug Discovery"},{"key":"2026100610484031900_cit4","doi-asserted-by":"publisher","first-page":"280","DOI":"10.1161\/01.RES.0000087541.15600.2B","article-title":"Cyclic GMP Phosphodiesterases and\nRegulation of Smooth Muscle Function","volume":"93","author":"Rybalkin","year":"2003","journal-title":"Circ.\nRes."},{"key":"2026100610484031900_cit5","doi-asserted-by":"publisher","first-page":"2554","DOI":"10.1111\/bph.14205","article-title":"PDE5 inhibitors\n\u2013 pharmacology and clinical applications 20 years after sildenafil\ndiscovery","volume":"175","author":"Andersson","year":"2018","journal-title":"Br. J. Pharmacol."},{"key":"2026100610484031900_cit6","doi-asserted-by":"publisher","first-page":"170","DOI":"10.1016\/j.sxmr.2016.02.004","article-title":"Non-Sexual\nImplications of Phosphodiesterase Type 5 Inhibitors","volume":"5","author":"Mostafa","year":"2017","journal-title":"Sex. Med. Rev."},{"key":"2026100610484031900_cit7","doi-asserted-by":"publisher","first-page":"246","DOI":"10.1177\/0976500X221129008","article-title":"A Review on the Pharmacological\nImportance\nof PDE5 and Its Inhibition to Manage Biomedical Conditions","volume":"13","author":"Saikia","year":"2022","journal-title":"J. Pharmacol. Pharmacother."},{"key":"2026100610484031900_cit8","doi-asserted-by":"publisher","first-page":"239","DOI":"10.1161\/01.CIR.0000081166.87607.E2","article-title":"Therapeutic potential\nof phosphodiesterase 5 inhibition\nfor cardiovascular disease","volume":"108","author":"Reffelmann","year":"2003","journal-title":"Circulation"},{"key":"2026100610484031900_cit9","doi-asserted-by":"publisher","first-page":"1244","DOI":"10.1136\/heartjnl-2017-312865","article-title":"Phosphodiesterase-5\ninhibitors and the heart: compound cardioprotection?","volume":"104","author":"Hutchings","year":"2018","journal-title":"Heart"},{"key":"2026100610484031900_cit10","doi-asserted-by":"publisher","DOI":"10.3390\/ph16091266","article-title":"Advancements in\nPhosphodiesterase 5 Inhibitors: Unveiling Present and Future Perspectives","volume":"16","author":"ElHady","year":"2023","journal-title":"Pharmaceuticals"},{"key":"2026100610484031900_cit11","doi-asserted-by":"publisher","first-page":"99179","DOI":"10.18632\/oncotarget.21837","article-title":"Phosphodiesterase type 5 and cancers:\nprogress and\nchallenges","volume":"8","author":"Barone","year":"2017","journal-title":"Oncotarget"},{"key":"2026100610484031900_cit12","doi-asserted-by":"publisher","first-page":"74","DOI":"10.1016\/j.sxmr.2015.10.004","article-title":"The\nRole of PDE5 Inhibitors and the NO\/cGMP Pathway in Cancer","volume":"4","author":"Peak","year":"2016","journal-title":"Sex. Med. Rev."},{"key":"2026100610484031900_cit13","doi-asserted-by":"publisher","DOI":"10.3332\/ecancer.2018.824","article-title":"Repurposing drugs in oncology (ReDO)-selective\nPDE5\ninhibitors as anti-cancer agents","volume":"12","author":"Pantziarka","year":"2018","journal-title":"Ecancermedicalscience"},{"key":"2026100610484031900_cit14","doi-asserted-by":"publisher","DOI":"10.3389\/fonc.2018.00681","article-title":"Phosphodiesterase Type\n5 (PDE5) Inhibitors Sensitize\nTopoisomerase II Inhibitors in Killing Prostate Cancer Through PDE5-Independent\nImpairment of HR and NHEJ DNA Repair Systems","volume":"8","author":"Chang","year":"2019","journal-title":"Front. Oncol."},{"key":"2026100610484031900_cit15","doi-asserted-by":"publisher","first-page":"832","DOI":"10.1021\/cn3000907","article-title":"Phosphodiesterases as Therapeutic Targets for Alzheimer\u2019s\nDisease","volume":"3","author":"Garci\u0301a-Osta","year":"2012","journal-title":"ACS Chem. Neurosci."},{"key":"2026100610484031900_cit16","doi-asserted-by":"publisher","first-page":"371","DOI":"10.2147\/NDT.S2447","article-title":"Role of phosphodiesterase\n5 in synaptic plasticity and memory","volume":"4","author":"Puzzo","year":"2008","journal-title":"Neuropsychiatr.\nDis. Treat."},{"key":"2026100610484031900_cit17","doi-asserted-by":"publisher","first-page":"8075","DOI":"10.1523\/JNEUROSCI.0864-09.2009","article-title":"Phosphodiesterase 5\nInhibition Improves Synaptic Function, Memory, and Amyloid- Load in\nan Alzheimer\u2019s Disease Mouse Model","volume":"29","author":"Puzzo","year":"2009","journal-title":"J.\nNeurosci."},{"key":"2026100610484031900_cit18","doi-asserted-by":"publisher","first-page":"585","DOI":"10.1007\/s10157-021-02051-6","article-title":"Reno-protective effects of Phosphodiesterase 5 inhibitors","volume":"25","author":"Coskuner","year":"2021","journal-title":"Clin. Exp. Nephrol."},{"key":"2026100610484031900_cit19","doi-asserted-by":"publisher","first-page":"1521","DOI":"10.1007\/s11255-015-1071-4","article-title":"Phosphodiesterase type\n5 inhibitors\nand kidney disease","volume":"47","author":"Afsar","year":"2015","journal-title":"Int. Urol. Nephrol."},{"key":"2026100610484031900_cit20","doi-asserted-by":"publisher","DOI":"10.3389\/fphar.2012.00167","article-title":"PDE5 Inhibitors\nas Potential Tools\nin the Treatment of Cystic Fibrosis","volume":"3","author":"Noel","year":"2012","journal-title":"Front.\nPharmacol."},{"key":"2026100610484031900_cit21","doi-asserted-by":"publisher","first-page":"H1486","DOI":"10.1152\/ajpheart.00301.2018","article-title":"Sildenafil improves\nvascular endothelial function in\npatients with cystic fibrosis","volume":"315","author":"Rodriguez-Miguelez","year":"2018","journal-title":"Am. J. Physiol.:\nHeart Circ. Physiol."},{"key":"2026100610484031900_cit22","doi-asserted-by":"publisher","first-page":"679","DOI":"10.1007\/s11010-022-04520-2","article-title":"Treating diabetes with combination\nof phosphodiesterase 5 inhibitors\nand hydroxychloroquine\u2500a possible prevention strategy for COVID-19?","volume":"478","author":"Kukreja","year":"2023","journal-title":"Mol. Cell. Biochem."},{"key":"2026100610484031900_cit23","doi-asserted-by":"publisher","first-page":"141","DOI":"10.1038\/s41573-023-00832-0","article-title":"Integrating QSAR modelling and deep learning in drug\ndiscovery: the emergence of deep QSAR","volume":"23","author":"Tropsha","year":"2024","journal-title":"Nat.\nRev. Drug Discovery"},{"key":"2026100610484031900_cit24","doi-asserted-by":"publisher","first-page":"346","DOI":"10.2174\/1389200215666140908102230","article-title":"A Review on Principles, Theory and Practices of 2D-QSAR","volume":"15","author":"Roy","year":"2014","journal-title":"Curr. Drug Metab."},{"key":"2026100610484031900_cit25","doi-asserted-by":"publisher","first-page":"3525","DOI":"10.1039\/D0CS00098A","article-title":"QSAR without\nborders","volume":"49","author":"Muratov","year":"2020","journal-title":"Chem. Soc. Rev."},{"key":"2026100610484031900_cit26","doi-asserted-by":"publisher","first-page":"505","DOI":"10.1002\/wcms.1187","article-title":"Modern 2D QSAR for\ndrug discovery","volume":"4","author":"Lewis","year":"2014","journal-title":"WIREs Comput.\nMol. Sci."},{"key":"2026100610484031900_cit27","doi-asserted-by":"publisher","first-page":"467","DOI":"10.1080\/13543776.2018.1475560","article-title":"QSAR modelling: a therapeutic patent\nreview 2010-present","volume":"28","author":"Halder","year":"2018","journal-title":"Expert Opin. Ther. Pat."},{"key":"2026100610484031900_cit28","doi-asserted-by":"publisher","first-page":"917","DOI":"10.1080\/1062936X.2021.1989721","article-title":"First molecular\nmodelling report on tri-substituted pyrazolines as phosphodiesterase\n5 (PDE5) inhibitors through classical and machine learning based multi-QSAR\nanalysis","volume":"32","author":"Nandi","year":"2021","journal-title":"SAR QSAR Environ. Res."},{"key":"2026100610484031900_cit29","doi-asserted-by":"publisher","DOI":"10.1186\/s13020-022-00705-5","article-title":"Isolation,\nbioassay and 3D-QSAR analysis of 8-isopentenyl flavonoids\nfrom Epimedium sagittatum maxim. as PDE5A inhibitors","volume":"17","author":"Li","year":"2022","journal-title":"Chin. Med."},{"key":"2026100610484031900_cit30","doi-asserted-by":"publisher","first-page":"820","DOI":"10.1007\/s00044-019-02311-x","article-title":"3D-QSAR modeling of\nPhosphodiesterase-5 inhibitors:\nevaluation and comparison of the receptor- and ligand-based alignments","volume":"28","author":"Jiang","year":"2019","journal-title":"Med. Chem. Res."},{"key":"2026100610484031900_cit31","doi-asserted-by":"publisher","DOI":"10.1016\/j.chemosphere.2023.141086","article-title":"Interpretable attention-based\nmulti-encoder transformer based QSPR model for assessing toxicity\nand environmental impact of chemicals","volume":"350","author":"Kim","year":"2024","journal-title":"Chemosphere"},{"key":"2026100610484031900_cit32","doi-asserted-by":"publisher","DOI":"10.1186\/s13321-020-00423-w","article-title":"Transformer-CNN: Swiss knife for\nQSAR modeling and interpretation","volume":"12","author":"Karpov","year":"2020","journal-title":"J. Cheminf."},{"key":"2026100610484031900_cit33","first-page":"237","article-title":"DRAGON software:\nAn easy approach to molecular descriptor calculations","volume":"56","author":"Mauri","year":"2006","journal-title":"MATCH Commun. Math. Comput. Chem."},{"key":"2026100610484031900_cit34","doi-asserted-by":"publisher","DOI":"10.1186\/s13321-024-00830-3","article-title":"Effectiveness of molecular fingerprints\nfor exploring the chemical space of natural products","volume":"16","author":"Boldini","year":"2024","journal-title":"J. Cheminf."},{"key":"2026100610484031900_cit35","doi-asserted-by":"publisher","DOI":"10.1186\/s13321-020-00443-6","article-title":"QSAR-derived affinity\nfingerprints (part 1): fingerprint construction and modeling performance\nfor similarity searching, bioactivity classification and scaffold\nhopping","volume":"12","author":"S\u030ckuta","year":"2020","journal-title":"J. Cheminf."},{"key":"2026100610484031900_cit36","doi-asserted-by":"publisher","first-page":"5363","DOI":"10.1021\/acs.jpcb.4c01875","article-title":"Utility of the Morgan\nFingerprint in Structure-Based\nVirtual Ligand Screening","volume":"128","author":"Zhou","year":"2024","journal-title":"J. Phys. Chem. B"},{"key":"2026100610484031900_cit37","doi-asserted-by":"publisher","first-page":"742","DOI":"10.1021\/ci100050t","article-title":"Extended-Connectivity\nFingerprints","volume":"50","author":"Rogers","year":"2010","journal-title":"J. Chem.\nInf. Model."},{"key":"2026100610484031900_cit38","doi-asserted-by":"publisher","DOI":"10.1002\/minf.201800082","article-title":"PySpark and RDKit: Moving towards\nBig Data in Cheminformatics","volume":"38","author":"Lovric\u0301","year":"2019","journal-title":"Mol. Inf."},{"key":"2026100610484031900_cit39","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-024-77948-9","article-title":"Deciphering antioxidant\ninteractions\nvia data mining and RDKit","volume":"15","author":"Ayres","year":"2025","journal-title":"Sci. Rep."},{"key":"2026100610484031900_cit40","doi-asserted-by":"publisher","first-page":"1273","DOI":"10.1021\/ci010132r","article-title":"Reoptimization of\nMDL Keys for Use in Drug Discovery","volume":"42","author":"Durant","year":"2002","journal-title":"J. Chem.\nInf. Comput. Sci."},{"key":"2026100610484031900_cit41","doi-asserted-by":"publisher","DOI":"10.1186\/s13321-021-00506-2","article-title":"Analysis of the effects of related fingerprints on molecular similarity\nusing an eigenvalue entropy approach","volume":"13","author":"Kuwahara","year":"2021","journal-title":"J. Cheminf."},{"key":"2026100610484031900_cit42","doi-asserted-by":"publisher","first-page":"107","DOI":"10.1021\/c160017a018","article-title":"The Generation\nof a Unique Machine Description for Chemical Structures-A Technique\nDeveloped at Chemical Abstracts Service","volume":"5","author":"Morgan","year":"1965","journal-title":"J.\nChem. Doc."},{"key":"2026100610484031900_cit43","doi-asserted-by":"publisher","DOI":"10.1186\/s13321-020-00445-4","article-title":"One molecular fingerprint to rule\nthem all: drugs, biomolecules, and the metabolome","volume":"12","author":"Capecchi","year":"2020","journal-title":"J. Cheminf."},{"key":"2026100610484031900_cit44","doi-asserted-by":"publisher","first-page":"21","DOI":"10.1109\/TIT.1967.1053964","article-title":"Nearest neighbor pattern\nclassification","volume":"13","author":"Cover","year":"1967","journal-title":"IEEE\nTrans. Inf. Theory"},{"key":"2026100610484031900_cit45","doi-asserted-by":"crossref","DOI":"10.1201\/9781315139470","volume-title":"Classification And\nRegression Trees","author":"Breiman","year":"2017","edition":"1"},{"key":"2026100610484031900_cit46","doi-asserted-by":"publisher","first-page":"1189","DOI":"10.1214\/aos\/1013203451","article-title":"Greedy\nFunction Approximation: A Gradient Boosting Machine","volume":"29","author":"Friedman","year":"2001","journal-title":"Ann. Stat."},{"key":"2026100610484031900_cit47","doi-asserted-by":"publisher","first-page":"308","DOI":"10.1016\/j.proeng.2011.11.2647","article-title":"Overview of\nSupport Vector Machine\nin Modeling Machining Performances","volume":"24","author":"Deris","year":"2011","journal-title":"Procedia\nEng."},{"key":"2026100610484031900_cit48","doi-asserted-by":"publisher","first-page":"224","DOI":"10.1109\/72.655045","article-title":"Upper bounds on the number of hidden neurons in feedforward\nnetworks with arbitrary bounded nonlinear activation functions","volume":"9","author":"Guang-Bin","year":"1998","journal-title":"IEEE Trans. Neural Networks"},{"key":"2026100610484031900_cit49","doi-asserted-by":"publisher","DOI":"10.3389\/fninf.2014.00014","article-title":"Machine learning for neuroimaging with scikit-learn","volume":"8","author":"Abraham","year":"2014","journal-title":"Front. Neuroinf."},{"key":"2026100610484031900_cit50","doi-asserted-by":"publisher","DOI":"10.3390\/molecules28176379","article-title":"In\nSilico Modeling and Structural\nAnalysis of Soluble Epoxide Hydrolase Inhibitors for Enhanced Therapeutic\nDesign","volume":"28","author":"Sar","year":"2023","journal-title":"Molecules"},{"key":"2026100610484031900_cit51","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0291908","article-title":"Challenges\nin the real world use of classification accuracy metrics: From recall\nand precision to the Matthews correlation coefficient","volume":"18","author":"Foody","year":"2023","journal-title":"PLoS One"},{"key":"2026100610484031900_cit52","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0177678","article-title":"Optimal classifier\nfor imbalanced\ndata using Matthews Correlation Coefficient metric","volume":"12","author":"Boughorbel","year":"2017","journal-title":"PLoS One"},{"key":"2026100610484031900_cit53","doi-asserted-by":"publisher","DOI":"10.1186\/s13040-021-00244-z","article-title":"The Matthews correlation\ncoefficient\n(MCC) is more reliable than balanced accuracy, bookmaker informedness,\nand markedness in two-class confusion matrix evaluation","volume":"14","author":"Chicco","year":"2021","journal-title":"BioData Min."},{"key":"2026100610484031900_cit54","doi-asserted-by":"publisher","DOI":"10.1186\/s13040-023-00322-4","article-title":"The Matthews correlation\ncoefficient (MCC) should replace\nthe ROC AUC as the standard metric for assessing binary classification","volume":"16","author":"Chicco","year":"2023","journal-title":"BioData Min."},{"key":"2026100610484031900_cit55","doi-asserted-by":"publisher","DOI":"10.1016\/j.jbi.2023.104426","article-title":"A statistical comparison\nbetween Matthews correlation\ncoefficient (MCC), prevalence threshold, and Fowlkes-Mallows index","volume":"144","author":"Chicco","year":"2023","journal-title":"J. Biomed. Inf."},{"key":"2026100610484031900_cit56","doi-asserted-by":"publisher","DOI":"10.1186\/s12864-019-6413-7","article-title":"The advantages of the Matthews correlation\ncoefficient\n(MCC) over F1 score and accuracy in binary classification evaluation","volume":"21","author":"Chicco","year":"2020","journal-title":"BMC Genomics"},{"key":"2026100610484031900_cit57","doi-asserted-by":"publisher","first-page":"353","DOI":"10.2217\/nmt-2021-0013","article-title":"Receiver operating\ncharacteristic plot and area under\nthe curve with binary classifiers: pragmatic analysis of cognitive\nscreening instruments","volume":"11","author":"Mbizvo","year":"2021","journal-title":"Neurodegener. Dis. Manage."},{"key":"2026100610484031900_cit58","doi-asserted-by":"publisher","first-page":"1282","DOI":"10.1080\/07391102.2018.1456975","article-title":"Identifying natural\ncompounds as multi-target-directed ligands against Alzheimer\u2019s\ndisease: an in silico approach","volume":"37","author":"Ambure","year":"2019","journal-title":"J. Biomol. Struct.\nDyn."},{"key":"2026100610484031900_cit59","doi-asserted-by":"publisher","first-page":"619","DOI":"10.1080\/1062936X.2023.2244419","article-title":"HDAC6 detector: online application for evaluating compounds\nas potential histone deacetylase 6 inhibitors","volume":"34","author":"Tinkov","year":"2023","journal-title":"SAR QSAR Environ. Res."},{"key":"2026100610484031900_cit60","doi-asserted-by":"publisher","DOI":"10.12688\/f1000research.8357.1","article-title":"Activity-relevant\nsimilarity values\nfor fingerprints and implications for similarity searching","volume":"5","author":"Jasial","year":"2016","journal-title":"F1000Research"},{"key":"2026100610484031900_cit61","doi-asserted-by":"publisher","DOI":"10.1186\/s40537-024-00905-w","article-title":"Feature\nselection strategies: a comparative analysis of SHAP-value and importance-based\nmethods","volume":"11","author":"Wang","year":"2024","journal-title":"J. Big Data"},{"key":"2026100610484031900_cit62","doi-asserted-by":"publisher","first-page":"8761","DOI":"10.1021\/acs.jmedchem.9b01101","article-title":"Interpretation of Compound\nActivity Predictions from\nComplex Machine Learning Models Using Local Approximations and Shapley\nValues","volume":"63","author":"Rodri\u0301guez-Pe\u0301rez","year":"2020","journal-title":"J. Med. Chem."},{"key":"2026100610484031900_cit63","doi-asserted-by":"publisher","first-page":"225","DOI":"10.1021\/ci00063a011","article-title":"Molecular\nidentification number for substructure searches","volume":"29","author":"Burden","year":"1989","journal-title":"J. Chem. Inf. Comput. Sci."}],"container-title":["ACS Omega"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/pubs.acs.org\/doi\/pdf\/10.1021\/acsomega.6c03768","content-type":"application\/pdf","content-version":"vor","intended-application":"unspecified"},{"URL":"https:\/\/pubs.acs.org\/acsodf\/article-pdf\/11\/39\/58896\/68567535\/acsomega.6c03768.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/pubs.acs.org\/acsodf\/article-pdf\/11\/39\/58896\/68567535\/acsomega.6c03768.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,6]],"date-time":"2026-10-06T15:29:58Z","timestamp":1791300598000},"score":1,"resource":{"primary":{"URL":"https:\/\/pubs.acs.org\/acsodf\/article\/11\/39\/58896\/5433489\/Interpretable-Machine-Learning-Frameworks-for-QSAR"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,20]]},"references-count":63,"journal-issue":{"issue":"39","published-online":{"date-parts":[[2026,9,20]]},"published-print":{"date-parts":[[2026,10,6]]}},"URL":"https:\/\/doi.org\/10.1021\/acsomega.6c03768","relation":{},"ISSN":["2470-1343"],"issn-type":[{"value":"2470-1343","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,20]]}}}