Citation:

Edwards, S., M. Nelms, AND N. Oki. Computational Approaches for Identifying Adverse Outcome Pathways. Environmental Mutagenesis & Genomics Society Annual Meeting, Raleigh, NC, September 09, 2002 - September 13, 2017.

Impact/Purpose:

Adverse Outcome Pathways can expand and enhance the use of ToxCast and other in vitro toxicity information by providing a mechanistic link to adverse outcomes of regulatory concern, but the expert-driven development of AOPs is labor-intensive and time-consuming. This work is intended to complement that process by creating a broad array of computationally-predicted AOPs (cpAOPs) by data mining of publicly available data. This increases the coverage of AOPs and provides minimal information in many cases where nothing is available otherwise. It also provides a starting point for expert-driven development of AOPs and thereby potentially accelerating that process.

Description:

Adverse Outcome Pathways (AOPs) provide a framework for organizing toxicity information to improve predictions of the potential adverse impact of environment stressors on humans or wildlife populations, but these benefits are currently limited by the small number of AOPs currently available. This talk will outline a process by which results from toxicogenomics and high throughput toxicity testing have been integrated with more traditional toxicological endpoints to programmatically assemble computationally predicted AOPs (cpAOPs). The process begins by building a global cpAOP network and then extracting the relevant nodes from this network based on an adverse outcome or high throughput assay target of interest. Building on previous efforts using frequent itemset mining to create the global cpAOP networks, automated extraction of subnetworks was carried out for liver related diseases. In addition to expert review of the resulting subnetworks, graph-based comparisons were done to evaluate the method. In the future, this information can be used by experts to develop novel AOPs, and the results of the expert evaluation can be used to refine the original computational predictions. A case study looking at the iterative development of cpAOPs and chemical groups will show how the development and use of AOPs can be coupled to provide short-term value while building a large knowledgebase of AOPs to provide more comprehensive decision support in the future. [This is an abstract or a proposed presentation and does not necessarily reflect EPA policy. Mention of trade names or commercial products does not constitute endorsement or recommendation for use.]

Record Details: