Machine Learning-Based Classification of Pairwise Selectivity Among Human Carbonic Anhydrase I, II, IX, and XII Isoenzymes
- 1Vocational School of Health Services, Antalya Bilim University, Antalya 07190, Turkey
- 2Department of Biology, Faculty of Science, Kahramanmaras Sutcu Imam University, Kahramanmaras 46040, Turkey
- 3Department of Mathematics and Science Teaching, Faculty of Education, Akdeniz University, Antalya 07070, Turkey
Abstract
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 ≤ 200 nM) and the paired isoenzyme was weakly inhibited (Ki > 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 ≈ 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.
Related
- Engineering cyanobacterial carbonic anhydrase surface mutations for fast CO2 hydration and capture
- Engineering cyanobacterial carbonic anhydrase surface mutations for fast CO2 hydration and capture
- Design, Synthesis, and Biological Evaluation of Novel Dithiocarbamate-Based Human Carbonic Anhydrase IX/XII Inhibitors as Potential Anti-Glioblastoma Agents
- Inhibitory, Anticancer Efficacy and Structural Insights of a Novel Ethyl 1-Hydroxy-3-Oxo-1-Phenylisoindoline-2-Sulfonate Compound Selectively Targeting Human CA IX and CA XII Carbonic Anhydrase, Ferroptosis-Mediated Cell Death.
- Interpretable Deep Learning of Endogenous Fluorophore Concentration and Entropy for Lung Tissue Classification
BibTeX
@article{hacisuleyman2026machine,
title = {{Machine Learning-Based Classification of Pairwise Selectivity Among Human Carbonic Anhydrase I, II, IX, and XII Isoenzymes}},
author = {Elif Gozde Hacisuleyman and Ahmet Kayraldiz and Erol Eroglu},
journal = {International Journal of Molecular Sciences},
year = {2026},
month = sep,
doi = {10.3390/ijms27198784},
url = {https://www.mdpi.com/1422-0067/27/19/8784},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Machine Learning-Based Classification of Pairwise Selectivity Among Human Carbonic Anhydrase I, II, IX, and XII Isoenzymes
AU - Elif Gozde Hacisuleyman
AU - Ahmet Kayraldiz
AU - Erol Eroglu
PY - 2026
DA - 2026/09/30
JO - International Journal of Molecular Sciences
DO - 10.3390/ijms27198784
UR - https://www.mdpi.com/1422-0067/27/19/8784
AB - 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 ≤ 200 nM) and the paired isoenzyme was weakly inhibited (Ki > 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 ≈ 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.
SN - 1422-0067
ER -