Fig. 1.

Schematic overview of the methods used to quantify the geospatial risk of molecular perturbation. The spatially explicit external chemical concentrations (C_ext(s)) are converted to an internal dose (D_int(s)) via inhalation parameters and using county-specific distributions of age and obesity prevalence (P(s)). D_int(s) is then converted into an in vitro equivalent steady-state plasma concentration (C_plasma(s)) using a steady-state plasma conversion factor (C_ss(s)) estimated using physiological based toxicokinetic (PBTK) modeling. The dashed box indicates that P(s) influences both the inhalation rate and C_ss(s) used to calculate C_plasma(s). The geospatial risk assessment R(s) is developed by using the C_plasma(s) and concentration-response models generated using curated high-throughput (cHTS) data, which are integrated using mixtures models (e.g., concentration addition) to predict the risk of exposure. We use a probabilistic framework using a Monte Carlo uncertainty analysis. The mapped risk of molecular perturbation can be integrated with the Adverse Outcome Pathway (AOP) framework to link the molecular-level perturbation to an adverse outcome at the individual and population levels.