In silico discovery of selective TPX2 and BUB1B inhibitors as novel antimitotic agents in breast cancer therapy.

Breast cancer (BC) is one of the most common malignancies in women globally, characterized by significant genetic and clinical heterogeneity. This complexity emphasises the need for reliable biomarkers and novel therapeutic strategies to improve patient outcomes.To address this, the present study introduces an integrated computational pipeline using multiple datasets to identify robust biomarkers and potential repurposed drugs in BC. From transcriptomic profiles across 12 GEO datasets, 143 differentially expressed genes (DEGs) were identified. Gene Ontology and functional enrichment analyses were performed, followed by the construction of a protein-protein interaction (PPI) network to pinpoint hub genes. These hubs were prioritised using multiple topological centrality measures and validated with independent datasets (cBioPortal, GEPIA, KM plotter) and machine-learning classification using five algorithms: Random Forest, XGBoost, Support vector machine, K-nearest neighbour, and Logistic regression. Machine-learning models achieved test accuracies > 0.90 on TCGA-BRCA data (n = 1203), with K-nearest neighbour (class-weighted: accuracy 0.954) and XGBoost (SMOTE: accuracy 0.931) showing the strongest performance. Prioritised hub genes, TPX2 and BUB1B, were subjected to virtual-screening across five drug databases (DrugBank, DGIdb, OpenTargets, SwissTargetPrediction, and GSCALite), combined with molecular-docking, ADMET profiling, and drug-likeness evaluation, to identify promising repurposed candidates. Vorinostat exhibited the highest binding affinity to TPX2 (-29.32 kcal/mol) and BUB1B (-23.71 kcal/mol), followed by BRD-K90370028 (-18.40 kcal/mol on BUB1B), NSC19630 and Dasatinib (with consistent dual-target binding), CD-437, and four additional prioritised compounds that exhibited favourable interactions. In conclusion, this coherent transcriptomics-to-therapeutics workflow establishes TPX2 and BUB1B as strong prognostic biomarkers in BC, with promising repurposed drugs targeting these mitotic regulators.