Session 2: Ontologies Hazard ID

Using Ontologies to Automate Hazard Identification From High Content and High Throughput Screening Assays

Lyle D. Burgoon, Chief,
Hazardous Pollutant Assessment Group


US Environmental Protection Agency (US EPA)


Today, there are over 80,000 chemicals in commerce or the environment, most of which lack authoritative toxicity information.

The ability to rapidly screen these chemicals for their potential hazard will allow risk assessors and risk managers to prioritize chemicals for further research and assessment.

Ontologies provide an efficient platform to integrate evidence for rapid hazard identification.

In this talk, I will demonstrate how we can utilize adverse outcome pathways (AOPs), ontologies, first order logic, and Bayesian evidence integration to make hazard hypotheses in support of rapid hazard identification based upon high content and high throughput screening data.

I will also discuss the potential for this process and the information generated by it to be automated into an expert system.

Disclaimer: the views expressed in this manuscript are those of the author and do not necessarily reflect the views or policies of the U.S. EPA.