A systems toxicology approach integrating network pharmacology, machine learning, and molecular docking reveals multi-target mechanisms of organophosphate pesticides in promoting breast cancer progression

Abstract Organophosphate pesticides (OPs) are widespread environmental contaminants implicated in breast cancer (BC) risk, yet their multi-target mechanisms remain unclear. This study integrated network toxicology, machine learning, and molecular docking to elucidate the biological links between OP exposure and BC. Targets of three OPs—parathion, malathion, and chlorpyrifos—and BC were collected from public databases, and overlapping genes were identified. Protein–protein interaction and enrichment analyses revealed these shared targets were predominantly enriched in cancer-related pathways, including MAPK and FoxO signaling. A LASSO-RF predictive model demonstrated excellent performance (C-index = 0.961) and identified a five-gene signature (F2, FABP4, EGFR, MMP13, and PPARG) significantly dysregulated in BC tissues. Molecular docking confirmed stable high-affinity binding between OPs and these targets. To further validate dynamic stability, 100 ns molecular dynamics simulations were performed for EGFR–chlorpyrifos, F2–parathion, and FABP4–malathion complexes, which showed stable binding with low RMSD and RMSF values; MM-PBSA calculations confirmed strong affinities (EGFR–chlorpyrifos: −17.89 kcal/mol). Collectively, our findings suggest OPs may promote BC through coordinated disruption of oncogenic and metabolic regulators. This study provides a systems-level framework for environmental carcinogenic risk assessment and highlights potential biomarkers linking OP exposure to BC.