{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,14]],"date-time":"2026-10-14T16:53:11Z","timestamp":1791996791470,"version":"4.4.0"},"reference-count":42,"publisher":"MDPI AG","issue":"19","license":[{"start":{"date-parts":[[2026,9,30]],"date-time":"2026-09-30T00:00:00Z","timestamp":1790726400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"},{"start":{"date-parts":[[2026,9,30]],"date-time":"2026-09-30T00:00:00Z","timestamp":1790726400000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Int. J. Mol. Sci.","IJMS"],"abstract":"<jats:p>Human carbonic anhydrase (hCA) isoenzymes participate in essential pH-regulatory processes, but conservation of their catalytic sites makes selective inhibition difficult. We developed a direct pairwise machine learning framework for selectivity among hCA I, II, IX, and XII using curated Ki data from ChEMBL. Compounds entered a binary task only when one isoenzyme was strongly inhibited (Ki \u2264 200 nM) and the paired isoenzyme was weakly inhibited (Ki &gt; 800 nM). A provenance-conservative STRICT analysis and a BROAD sensitivity analysis were evaluated under train-only split selection, hyperparameter optimization, and a globally locked outer-test protocol. The highest observed held-out performance was obtained for CA II\/CA XII (MCC = 0.748) and CA IX\/CA XII (MCC = 0.718), whereas CA II\/CA IX was intermediate (MCC = 0.459) and CA-I-containing tasks were generally more difficult. Because the highest point estimates arose from relatively small test cohorts, these findings should be regarded as promising and require independent external or prospective validation. All six STRICT models exceeded 100 randomized-label controls (empirical p = 1\/101 \u2248 0.0099, the resolution floor for 100 permutations). Applicability domain and SHAP analyses indicated pair-specific signals compatible with known isoenzyme-dependent active-site differences. Overall, the framework prioritizes interpretable hCA selectivity hypotheses while explicitly defining current uncertainty and validation limits.<\/jats:p>","DOI":"10.3390\/ijms27198784","type":"journal-article","created":{"date-parts":[[2026,9,30]],"date-time":"2026-09-30T15:20:59Z","timestamp":1790781659000},"page":"8784","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Machine Learning-Based Classification of Pairwise Selectivity Among Human Carbonic Anhydrase I, II, IX, and XII Isoenzymes"],"prefix":"10.3390","volume":"27","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-8533-4023","authenticated-orcid":false,"given":"Elif Gozde","family":"Hacisuleyman","sequence":"first","affiliation":[{"name":"Vocational School of Health Services, Antalya Bilim University, Antalya 07190, Turkey"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ahmet","family":"Kayraldiz","sequence":"additional","affiliation":[{"name":"Department of Biology, Faculty of Science, Kahramanmaras Sutcu Imam University, Kahramanmaras 46040, Turkey"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5401-987X","authenticated-orcid":false,"given":"Erol","family":"Eroglu","sequence":"additional","affiliation":[{"name":"Department of Mathematics and Science Teaching, Faculty of Education, Akdeniz University, Antalya 07070, Turkey"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,9,30]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","unstructured":"Occhipinti, R., and Boron, W.F. (2019). Role of Carbonic Anhydrases and Inhibitors in Acid-Base Physiology: Insights from Mathematical Modeling. Int. J. Mol. Sci., 20.","DOI":"10.3390\/ijms20153841"},{"key":"ref_2","doi-asserted-by":"crossref","unstructured":"Supuran, C.T. (2023). Carbonic anhydrase versatility: From pH regulation to CO2 sensing and metabolism. Front. Mol. Biosci., 10.","DOI":"10.3389\/fmolb.2023.1326633"},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"150","DOI":"10.1021\/acs.chemrev.4c00278","article-title":"Multiple Binding Modes of Inhibitors to Human Carbonic Anhydrases: An Update on the Design of Isoform-Specific Modulators of Activity","volume":"125","author":"Alterio","year":"2025","journal-title":"Chem. Rev."},{"key":"ref_4","doi-asserted-by":"crossref","unstructured":"S\u00fcmer, A., \u015eahin, S., and Mente\u015fe, A. (2026). Carbonic Anhydrase I and II as Biomarkers and Therapeutic Targets in Human Disease: From Physiology to Clinical Translation. Int. J. Mol. Sci., 27.","DOI":"10.3390\/ijms27146375"},{"key":"ref_5","doi-asserted-by":"crossref","unstructured":"Singh, S., Lomelino, C.L., Mboge, M.Y., Frost, S.C., and McKenna, R. (2018). Cancer Drug Development of Carbonic Anhydrase Inhibitors beyond the Active Site. Molecules, 23.","DOI":"10.3390\/molecules23051045"},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"358","DOI":"10.1158\/0008-5472.CAN-08-2470","article-title":"Hypoxia-inducible carbonic anhydrase IX and XII promote tumor cell growth by counteracting acidosis through the regulation of the intracellular pH","volume":"69","author":"Chiche","year":"2009","journal-title":"Cancer Res."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"112923","DOI":"10.1016\/j.ejmech.2020.112923","article-title":"Recent advances in the medicinal chemistry of carbonic anhydrase inhibitors","volume":"209","author":"Kumar","year":"2021","journal-title":"Eur. J. Med. Chem."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"341","DOI":"10.1111\/cbdd.12759","article-title":"Quantitative Characterization of the Interaction Space of the Mammalian Carbonic Anhydrase Isoforms I, II, VII, IX, XII, and XIV and their Inhibitors, Using the Proteochemometric Approach","volume":"88","author":"Rasti","year":"2016","journal-title":"Chem. Biol. Drug Des."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"270","DOI":"10.1016\/j.molstruc.2018.02.107","article-title":"Development of selective QSAR models and molecular docking study for inhibitory activity of sulfonamide derivatives against carbonic anhydrase isoforms II and IX","volume":"1163","author":"Yorulmaz","year":"2018","journal-title":"J. Mol. Struct."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"234","DOI":"10.1016\/j.compbiolchem.2019.03.005","article-title":"A selectivity study of benzenesulfonamide derivatives on human carbonic anhydrase II\/IX by 3D-QSAR, Molecular Docking and Molecular Dynamics Simulation","volume":"80","author":"Wang","year":"2019","journal-title":"Comput. Biol. Chem."},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"307","DOI":"10.1016\/j.compbiolchem.2019.04.011","article-title":"DFT-based QSAR modelling of selectivity and inhibitory activity of coumarins and sulfocoumarins against tumor-associated carbonic anhydrase isoform IX","volume":"80","author":"Eroglu","year":"2019","journal-title":"Comput. Biol. Chem."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"127844","DOI":"10.1016\/j.molstruc.2020.127844","article-title":"DFT based QSARs for inhibitory activity of coumarins towards tumor-associated isoform (CA XII) of carbonic anhydrases","volume":"1208","author":"Yorulmaz","year":"2020","journal-title":"J. Mol. Struct."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"365","DOI":"10.1080\/14756366.2019.1705291","article-title":"Development of a cheminformatics platform for selectivity analyses of carbonic anhydrase inhibitors","volume":"35","author":"Poli","year":"2020","journal-title":"J. Enzym. Inhib. Med. Chem."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"18","DOI":"10.1186\/s13321-021-00499-y","article-title":"Prediction of activity and selectivity profiles of human Carbonic Anhydrase inhibitors using machine learning classification models","volume":"13","author":"Tinivella","year":"2021","journal-title":"J. Cheminform."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"4080","DOI":"10.1021\/acsomega.0c06153","article-title":"Predicting Isoform-Selective Carbonic Anhydrase Inhibitors via Machine Learning and Rationalizing Structural Features Important for Selectivity","volume":"6","author":"Galati","year":"2021","journal-title":"ACS Omega"},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Piazza, L., Di Stefano, M., Poles, C., Bononi, G., Poli, G., Renzi, G., Galati, S., Giordano, A., Macchia, M., and Carta, F. (2025). A Machine Learning Platform for Isoform-Specific Identification and Profiling of Human Carbonic Anhydrase Inhibitors. Pharmaceuticals, 18.","DOI":"10.3390\/ph18071007"},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"8419","DOI":"10.1038\/s41598-026-39771-2","article-title":"Interpretable machine learning rationalizes carbonic anhydrase inhibition via conformal and counterfactual prediction","volume":"16","author":"Ghamsary","year":"2026","journal-title":"Sci. Rep."},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"2270180","DOI":"10.1080\/14756366.2023.2270180","article-title":"Design, synthesis, and biological investigation of selective human carbonic anhydrase II, IX, and XII inhibitors using 7-aryl\/heteroaryl triazolopyrimidines bearing a sulfanilamide scaffold","volume":"38","author":"Romagnoli","year":"2023","journal-title":"J. Enzym. Inhib. Med. Chem."},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"681","DOI":"10.1002\/ddr.22049","article-title":"Selective Carbonic Anhydrase IX and XII Inhibitors Based Around a Functionalized Coumarin Scaffold","volume":"84","author":"Huwaimel","year":"2023","journal-title":"Drug Dev. Res."},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"298","DOI":"10.1006\/jmbi.1994.1655","article-title":"Drug-protein interactions. Refined structures of three sulfonamide drug complexes of human carbonic anhydrase I enzyme","volume":"243","author":"Chakravarty","year":"1994","journal-title":"J. Mol. Biol."},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"274","DOI":"10.1002\/prot.340040406","article-title":"Refined structure of human carbonic anhydrase II at 2.0 \u00c5 resolution","volume":"4","author":"Eriksson","year":"1988","journal-title":"Proteins"},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"16233","DOI":"10.1073\/pnas.0908301106","article-title":"Crystal structure of the catalytic domain of the tumor-associated human carbonic anhydrase IX","volume":"106","author":"Alterio","year":"2009","journal-title":"Proc. Natl. Acad. Sci. USA"},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"9545","DOI":"10.1073\/pnas.161301298","article-title":"Crystal structure of the dimeric extracellular domain of human carbonic anhydrase XII, a bitopic membrane protein overexpressed in certain cancer tumor cells","volume":"98","author":"Whittington","year":"2001","journal-title":"Proc. Natl. Acad. Sci. USA"},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"453543","DOI":"10.1155\/2015\/453543","article-title":"Probing the Surface of Human Carbonic Anhydrase for Clues towards the Design of Isoform Specific Inhibitors","volume":"2015","author":"Pinard","year":"2015","journal-title":"BioMed Res. Int."},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"110682","DOI":"10.1016\/j.abb.2025.110682","article-title":"A comprehensive review of the structural basis for the selectivity of sulfonamide-based inhibitors for carbonic anhydrase isoforms IX and XII over carbonic anhydrase II based on the available high quality crystallographic data","volume":"777","author":"Ahmad","year":"2026","journal-title":"Arch. Biochem. Biophys."},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"184","DOI":"10.1016\/j.ejmech.2017.03.026","article-title":"Structure activity study of carbonic anhydrase IX: Selective inhibition with ureido-substituted benzenesulfonamides","volume":"132","author":"Mboge","year":"2017","journal-title":"Eur. J. Med. Chem."},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"D1180","DOI":"10.1093\/nar\/gkad1004","article-title":"The ChEMBL Database in 2023: A drug discovery platform spanning multiple bioactivity data types and time periods","volume":"52","author":"Zdrazil","year":"2024","journal-title":"Nucleic Acids Res."},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"3","DOI":"10.1016\/S0169-409X(00)00129-0","article-title":"Experimental and computational approaches to estimate solubility and permeability in drug discovery and development settings","volume":"46","author":"Lipinski","year":"2001","journal-title":"Adv. Drug Deliv. Rev."},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"2887","DOI":"10.1021\/jm9602928","article-title":"The properties of known drugs. 1. Molecular frameworks","volume":"39","author":"Bemis","year":"1996","journal-title":"J. Med. Chem."},{"key":"ref_30","unstructured":"Landrum, G. (2014). RDKit: Open-Source Cheminformatics. Release 2014.03.1, Zenodo."},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"51","DOI":"10.1186\/s13321-020-00456-1","article-title":"An open source chemical structure curation pipeline using RDKit","volume":"12","author":"Bento","year":"2020","journal-title":"J. Cheminform."},{"key":"ref_32","doi-asserted-by":"crossref","first-page":"4","DOI":"10.1186\/s13321-018-0258-y","article-title":"Mordred: A molecular descriptor calculator","volume":"10","author":"Moriwaki","year":"2018","journal-title":"J. Cheminform."},{"key":"ref_33","doi-asserted-by":"crossref","first-page":"742","DOI":"10.1021\/ci100050t","article-title":"Extended-Connectivity Fingerprints","volume":"50","author":"Rogers","year":"2010","journal-title":"J. Chem. Inf. Model."},{"key":"ref_34","doi-asserted-by":"crossref","first-page":"137","DOI":"10.1080\/00401706.1969.10490666","article-title":"Computer Aided Design of Experiments","volume":"11","author":"Kennard","year":"1969","journal-title":"Technometrics"},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"273","DOI":"10.1023\/A:1022627411411","article-title":"Support-Vector Networks","volume":"20","author":"Cortes","year":"1995","journal-title":"Mach. Learn."},{"key":"ref_36","first-page":"2825","article-title":"Scikit-learn: Machine Learning in Python","volume":"12","author":"Pedregosa","year":"2011","journal-title":"J. Mach. Learn. Res."},{"key":"ref_37","doi-asserted-by":"crossref","unstructured":"Chicco, D., and Jurman, G. (2020). The advantages of the Matthews correlation coefficient (MCC) over F1 score and accuracy in binary classification evaluation. BMC Genom., 21.","DOI":"10.1186\/s12864-019-6413-7"},{"key":"ref_38","doi-asserted-by":"crossref","unstructured":"Saito, T., and Rehmsmeier, M. (2015). The Precision-Recall Plot Is More Informative than the ROC Plot When Evaluating Binary Classifiers on Imbalanced Datasets. PLoS ONE, 10.","DOI":"10.1371\/journal.pone.0118432"},{"key":"ref_39","doi-asserted-by":"crossref","first-page":"4791","DOI":"10.3390\/molecules17054791","article-title":"Comparison of Different Approaches to Define the Applicability Domain of QSAR Models","volume":"17","author":"Sahigara","year":"2012","journal-title":"Molecules"},{"key":"ref_40","doi-asserted-by":"crossref","first-page":"694","DOI":"10.1002\/qsar.200610151","article-title":"Principles of QSAR models validation: Internal and external","volume":"26","author":"Gramatica","year":"2007","journal-title":"QSAR Comb. Sci."},{"key":"ref_41","doi-asserted-by":"crossref","first-page":"2345","DOI":"10.1021\/ci700157b","article-title":"y-Randomization and Its Variants in QSPR\/QSAR","volume":"47","author":"Meringer","year":"2007","journal-title":"J. Chem. Inf. Model."},{"key":"ref_42","unstructured":"Lundberg, S.M., and Lee, S.-I. (2017). A Unified Approach to Interpreting Model Predictions. Advances in Neural Information Processing Systems 30, Curran Associates, Inc."}],"container-title":["International Journal of Molecular Sciences"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1422-0067\/27\/19\/8784\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,14]],"date-time":"2026-10-14T16:42:01Z","timestamp":1791996121000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1422-0067\/27\/19\/8784"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,30]]},"references-count":42,"journal-issue":{"issue":"19","published-online":{"date-parts":[[2026,10]]}},"alternative-id":["ijms27198784"],"URL":"https:\/\/doi.org\/10.3390\/ijms27198784","relation":{},"ISSN":["1422-0067"],"issn-type":[{"value":"1422-0067","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,30]]}}}