Data Interoperability in Toxicology: Supporting Source-to-Outcome Assessment

Hosted by the SOT Risk Assessment and Computational Toxicology Specialty Sections

Webinar Description:

Integrated approaches that leverage multiple data streams can provide strong, evidence-based support for risk assessment and decision making, but the practicality of data integration across Environmental Health Science (EHS) fields can be challenging because the current data landscape is largely fragmented. Domains such as toxicology, epidemiology, and exposure science have heterogeneous and inconsistent data formats, nuanced methodologies and data collection strategies, and varied reporting practices, limiting interoperability across disciplines. 

This work aims to improve data interoperability across EHS domains and demonstrate the value of integrated mechanistic analyses to inform decision making. We present a pilot use case that spans the Source-to-Outcome (S2O) continuum and focuses on PM₂.₅ exposure and decreased lung to exemplify the complexity of linking real-world environmental exposure data to mechanistic biological understanding and health impacts. The exposure component of the use case employs a synthetic population, air quality data, and variation in demographic and behavioral factors to simulate the Aggregate Exposure Pathway (AEP), influencing PM₂.₅ intake during wildfire events. The biological component leverages Adverse Outcome Pathway (AOP) networks to represent mechanistic relationships through AOP Bayesian Network (AOPBN) modeling. 

This methodology provides a probabilistic structure for integrating diverse evidence streams and evaluating the impacts of data integration on predictive confidence. Dosimetry modeling represents the critical bridge between exposure and outcome components. Our use case demonstrates how standardized, interoperable data ecosystems that span the S2O continuum can support comprehensive risk assessments and inform on human health outcomes.