Molecular Representation and Closed-Loop Validation for Toxicity Assessment of Organic Compounds in Ambient Air PM 2.5

Although the health impacts of fine particulate matter (PM_2.5) are primarily attributed to its chemically diverse composition rather than to mass concentration, assessing the toxicity of PM_2.5 constituent compounds remains highly challenging due to chemical complexity and limited experimental scalability. This study introduces an interpretable machine learning (ML) framework using a curated A549 cytotoxicity data set (19,841 compounds) that integrates customized Molecular Access System (MACCS) fingerprints, six base models, and a stacking ensemble meta-model, all optimized with a biobjective strategy to assess the toxicity of organic compounds in PM_2.5. The stacking ensemble model demonstrated satisfactory performance (test AUC > 0.8), exhibiting good generalization, adaptability, and robustness. Nontargeted analysis generated and experimentally validated a prediction set from the Hong Kong PM_2.5 samples (51 of 387 compounds confirmed from 13 classes), demonstrating broad applicability on an independent Nanjing PM_2.5 set (572 compounds). Key substructures driving toxicity, identified through a SHapley Additive exPlanations (SHAP) analysis and cell experimental validation, revealed that PM_2.5 compounds with aromatic rings and nitrogen-based functional groups (e.g., aromatic α,β-unsaturated ketones, aromatic amines, aromatic nitro compounds, and tertiary amines) likely contribute to high toxicity. The derived "structure-toxicity" rules narrow the search space from thousands of compounds to those containing critical substructures, enabling efficient prioritization of toxic components and providing a foundation for improving model specificity and predictive accuracy in future studies.