A Data-Driven Approach for the Development of a Time-Informed Adverse Outcome Pathway Network for the Cardiotoxicity of Environmental Chemicals

Abstract

We developed a bottom-up, data-driven approach for constructing a time-informed adverse outcome pathway (AOP) network for environmental chemical-induced cardiotoxicity. Mechanistic end points were systematically extracted from 339 in vitro and in vivo studies, yielding 1759 Key Event (KE) entries and 4938 Key Event Relationship (KER) entries, together with study metadata and essentiality evidence from intervention experiments. After quality filtering and minimum support thresholding, 62 unique KEs and 256 unique KERs were retained for network construction. Network analysis highlighted oxidative stress and mitochondrial dysfunction as prominent, highly connected KEs linking diverse upstream perturbations to downstream cardiomyocyte injury, inflammation, cardiac remodeling (fibrosis and hypertrophy), decreased myocardial contractility, and reduced left ventricular function. A central methodological finding was that temporal information was more informative at the level of KERs than at the level of KEs. The same KE could occur across multiple durations depending on the relationship in which it was embedded, whereas KER-level time-scale annotation resolved distinct acute, subacute, and chronic pathway patterns not apparent from KE timing alone. This framework provides an evidence-weighted synthesis of the most consistently supported cardiotoxicity mechanisms and introduces a temporal layer for the mechanistic interpretation of heterogeneous evidence.