Explainable Machine Learning for Prioritising Oxidative-Nucleotide and Cell-Cycle Targets in Ovarian Cancer
Keywords:
ovarian cancer, NUDT1, CDK4, machine learning, multi-omics, explainable AIAbstract
Background: Integrating genomic, transcriptomic and clinical data may identify biologically coherent therapeutic targets that are missed by single-layer analyses.
Objective: We developed an explainable machine-learning pipeline to assess whether oxidative-nucleotide defence and cell-cycle regulators contribute jointly to poor outcomes in high-grade serous ovarian cancer.
Methods: Publicly available tumour profiles from 498 patients were harmonised for gene expression, copy-number status, selected mutations and survival. A gradient-boosted survival model was compared with penalised Cox regression. SHAP values were used to quantify feature contribution, and nested resampling was applied to limit optimistic performance estimates.
Results: The boosted model achieved a time-dependent AUC of 0.72 at three years compared with 0.65 for the penalised Cox model. Elevated NUDT1 and CDK4 expression were consistently among the 20 highest-contributing molecular features, and their combined high-expression phenotype was associated with shorter median progression-free survival (14.2 versus 20.8 months in the analysis cohort). SHAP interaction analysis suggested that the adverse contribution of NUDT1 was amplified in tumours with high E2F-pathway activity. A reduced 12-feature model retained most of the predictive signal and showed better calibration than the full feature set. Sensitivity analyses excluding extreme copy-number events did not materially alter the ranking of NUDT1 or CDK4.
Conclusion: Explainable multi-omics modelling identified a reproducible interaction between oxidative nucleotide sanitisation and cell-cycle activation in ovarian cancer. The results justify targeted laboratory experiments and illustrate how transparent machine learning can support mechanism-oriented prioritisation rather than function solely as a black-box prognostic tool.
Publication format: research abstract. No full-text manuscript accompanies this record.
References
Miah R, Issahaku AR, Uddin MN, Hamali MA, Miah MM, Kassim K, Wilhelm A, Mohd Aluwi MFF. Benzimidazole Derivative as a Promising Targeted Therapeutic for Lung Cancer: In-Silico Approaches. J Pharm Innov. 2026;21:409. doi:10.1007/s12247-026-10615-w.
Tasnim J, Hasan MM, Hossain A, Miah MM, Hossain KR, Molla MTH, Ahmed Z, Reza AFGM, Alam MT. In silico identification of a potential NUDT1 inhibitor for cancer treatment. Sci Rep. 2026. doi:10.1038/s41598-026-66781-x.
Ibrahim HA, Refaat HM. Versatile mechanisms of 2-substituted benzimidazoles in targeted cancer therapy. Future J Pharm Sci. 2020;6(1):41.
Almalki AS, Nazreen S, Elbehairi SEI, Asad M, Shati AA, Alfaifi MY, Alam MM. Design, synthesis, anticancer activity and molecular docking studies of new benzimidazole derivatives bearing 1,3,4-oxadiazole moieties as potential thymidylate synthase inhibitors. New J Chem. 2022;46(31):14967-14978.
Youssif BG, Morcoss MM, Bräse S, Abdel-Aziz M, Abdel-Rahman HM, Abdelhafez ESM, et al. Benzimidazole-based derivatives as apoptotic antiproliferative agents: design, synthesis, docking, and mechanistic studies. Molecules. 2024;29(2):446.
Wagih N, Abdel-Rahman IM, El-Koussi NA, Abuo-Rahma GEDA. Anticancer benzimidazole derivatives as inhibitors of epigenetic targets: a review article. RSC Adv. 2025;15(2):966-1010.
Garg A, Khan MA. A systematic review of the impact of benzimidazole-based anthelmintics on lung cancer in animal models. Curr Drug Saf. 2025;20(3):303-310.
Nguyen VT, Huynh TKC, Ho GTT, Nguyen THA, Nguyen TLA, Dao DQ, et al. Metal complexes of benzimidazole-derived ligands as potential anticancer agents: synthesis, characterization, combined experimental and computational studies. R Soc Open Sci. 2022;9(9):220659.
Hussain A, AlAjmi MF, Rehman MT, Khan AA, Shaikh PA, Khan RA. Evaluation of transition metal complexes of benzimidazole-derived scaffold as promising anticancer chemotherapeutics. Molecules. 2018;23(5):1232.
Hopkins AL. Network pharmacology: the next paradigm in drug discovery. Nat Chem Biol. 2008;4(11):682-690.
Meng XY, Zhang HX, Mezei M, Cui M. Molecular docking: a powerful approach for structure-based drug discovery. Curr Comput Aided Drug Des. 2011;7(2):146-157.
Ferreira LG, dos Santos RN, Oliva G, Andricopulo AD. Molecular docking and structure-based drug design strategies. Molecules. 2015;20(7):13384-13421.
Hollingsworth SA, Dror RO. Molecular dynamics simulation for all. Neuron. 2018;99(6):1129-1143.
Hospital A, Goñi JR, Orozco M, Gelpí JL. Molecular dynamics simulations: advances and applications. Adv Appl Bioinform Chem. 2015;8:37-47.
Genheden S, Ryde U. The MM/PBSA and MM/GBSA methods to estimate ligand-binding affinities. Expert Opin Drug Discov. 2015;10(5):449-461.
Wang E, Sun H, Wang J, Wang Z, Liu H, Zhang JZH, Hou T. End-point binding free energy calculation with MM/PBSA and MM/GBSA: strategies and applications in drug design. Chem Rev. 2019;119(16):9478-9508.
Gfeller D, Grosdidier A, Wirth M, Daina A, Michielin O, Zoete V. SwissTargetPrediction: a web server for target prediction of bioactive small molecules. Nucleic Acids Res. 2014;42(W1):W32-W38.
Huang DW, Sherman BT, Lempicki RA. Bioinformatics enrichment tools: paths toward comprehensive functional analysis of large gene lists. Nucleic Acids Res. 2009;37(1):1-13.
Szklarczyk D, Gable AL, Lyon D, Junge A, Wyder S, Huerta-Cepas J, et al. STRING v11: protein-protein association networks with increased coverage, supporting functional discovery in genome-wide experimental datasets. Nucleic Acids Res. 2019;47(D1):D607-D613.
Shannon P, Markiel A, Ozier O, Baliga NS, Wang JT, Ramage D, et al. Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome Res. 2003;13(11):2498-2504.
Chin CH, Chen SH, Wu HH, Ho CW, Ko MT, Lin CY. cytoHubba: identifying hub objects and sub-networks from complex interactome. BMC Syst Biol. 2014;8(Suppl 4):S11.
Day PJ, Cleasby A, Tickle IJ, O'Reilly M, Coyle JE, Holding FP, et al. Crystal structure of human CDK4 in complex with a D-type cyclin. Proc Natl Acad Sci U S A. 2009;106(11):4166-4170.
Maier JA, Martinez C, Kasavajhala K, Wickstrom L, Hauser KE, Simmerling C. ff14SB: improving the accuracy of protein side chain and backbone parameters from ff99SB. J Chem Theory Comput. 2015;11(8):3696-3713.
Wang J, Wolf RM, Caldwell JW, Kollman PA, Case DA. Development and testing of a general AMBER force field. J Comput Chem. 2004;25(9):1157-1174.
Daina A, Michielin O, Zoete V. SwissADME: a free web tool to evaluate pharmacokinetics, drug-likeness and medicinal chemistry friendliness of small molecules. Sci Rep. 2017;7:42717.
Banerjee P, Kemmler E, Dunkel M, Preissner R. ProTox 3.0: a webserver for the prediction of toxicity of chemicals. Nucleic Acids Res. 2024;52(W1):W513-W520.
Rácz A, Mihalovits LM, Bajusz D, Héberger K, Miranda-Quintana RA. Molecular dynamics simulations and diversity selection by extended continuous similarity indices. J Chem Inf Model. 2022;62(14):3415-3425.
Moukengue B, Brown HK, Charrier C, Battaglia S, Baud'Huin M, Quillard T, et al. TH1579, MTH1 inhibitor, delays tumour growth and inhibits metastases development in osteosarcoma model. EBioMedicine. 2020;53:102704.
Zhou W, Ma L, Yang J, Qiao H, Li L, Guo Q, et al. Potent and specific MTH1 inhibitors targeting gastric cancer. Cell Death Dis. 2019;10(6):434.
Zhan D, Zhang X, Li J, Ding X, Cui Y, Jia J. MTH1 inhibitor TH287 suppresses gastric cancer development through the regulation of PI3K/AKT signaling. Cancer Biother Radiopharm. 2020;35(3):223-232.
Published
Issue
Section
License
Copyright (c) 2026 Farhan Kabir, Naledi Mokoena, Valeria Hernández
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors retain copyright. This work is licensed under a Creative Commons Attribution 4.0 International Licence (CC BY 4.0). Sharing and adaptation are permitted with appropriate attribution, a link to the licence and an indication of changes. Third-party material may be subject to separate terms stated in its credit line.
