An AI-based integrative framework with multi-omics and experimental validation reveals mechanisms of DABP-induced neurotoxicity.

4,4'-Diaminobenzophenone (DABP) is an emerging aromatic amine-benzophenone environmental contaminant with potential neurotoxic risk, while its molecular mechanisms remain insufficiently defined. Using an integrative framework combining network toxicology, machine learning-based target prioritization, molecular docking, in vitro neurotoxicity assays, and transcriptomic-metabolomic profiling, we systematically characterized the neurotoxic effects of DABP. Network analysis revealed prominent perturbations in mitogen-activated protein kinase (MAPK), phosphatidylinositol 3-kinase-protein kinase (PI3K-Akt), calcium signaling, and inflammatory pathways. Machine learning approaches consistently identified ciliary neurotrophic factor (CNTF), endothelin 1 (EDN1), semaphorin 3F (SEMA3F), growth arrest and DNA damage-inducible alpha (GADD45A), and FBJ murine osteosarcoma viral oncogene homolog (FOS) as core drivers of DABP-induced neurotoxicity, with SHapley Additive exPlanations (SHAP) enabling quantitative interpretation of their contributions. Molecular docking supported stable interactions between DABP and these targets. Functionally, DABP exposure induced oxidative stress, mitochondrial dysfunction, lipid peroxidation, intracellular iron accumulation, and ferroptosis-related processes in SH-SY5Y cells, accompanied by sustained neuroinflammatory activation. Integrated transcriptomic and metabolomic analyses revealed coordinated dysregulation of inflammatory signaling, metabolic homeostasis, and synaptic-associated pathways. Collectively, this study establishes a systems-level mechanistic framework for DABP-induced neurotoxicity, providing critical insights for environmental neurotoxicity risk assessment and mechanistic toxicology of aromatic amine pollutants.