Capturing Differential Quality of Experimental Evidence in a Predictive Quantum-Mechanical Model for Respiratory Sensitization.

Abstract

Asthma is of concern in occupational toxicology with significant public-health and economic costs. In the absence of benchmark in vivo and in vitro tests, the use of mechanistically sound in silico models is critical to inform hazard and to protect workers from exposure to potentially harmful substances. We recently reported on the computer-aided discovery and REdesign (CADRE) model for respiratory sensitization, which relies on a tiered structure of expert rules, molecular simulations, quantum-mechanics calculations and advanced statistics to accurately identify respiratory sensitizers from first principles. Here, we present an update to this model based on two years of testing in the pharmaceutical space, where we captured the heterogeneity of the underlying experimental evidence in two predictive tiers, thus allowing the practitioner to select an outcome based on their expert assessment of the data reliability and relevance. This user-based tuning of predictive models is critical for end points that lack consensus on what constitutes satisfactory evidence to support a decision in the handling of chemicals for occupational safety.