You are here:
Extracting and benchmarking emerging adverse outcome pathway knowledge
Citation:
Pollesch, N., Dan Villeneuve, AND J. O'Brien. Extracting and benchmarking emerging adverse outcome pathway knowledge. TOXICOLOGICAL SCIENCES. Society of Toxicology, RESTON, VA, 168(2):349-364, (2019). https://doi.org/10.1093/toxsci/kfz006
Impact/Purpose:
Linking molecular level interactions with chemicals or other stressors to adverse outcomes at higher levels of biological organization that are of regulatory significance is a challenge. Adverse outcome pathways (AOPs) provide a structured methodology to address this challenge. In addition to structuring toxicological knowledge, AOPs have found use in a variety of applications, from serving to facilitate weight of evidence approaches for risk assessment to providing a structure for quantitative modeling efforts (qAOPs). This research product is two-fold, first, it develops and introduces a set of mathematical and statistical techniques for analyzing AOP networks that allow researchers to extract additional information and value from their investment in constructing an AOP and contributing to the AOP knowledge base. Second, the benchmarking aspect of this product will provide an assessment of progress, useful for tracking AOP development and highlighting areas of interest to EPA researchers and beyond.
Description:
In order to link molecular level perturbations elicited by chemicals or other stressors to adverse outcomes at higher levels of biological organization that are typically of regulatory significance, a mechanistic description is often required. Adverse outcome pathways (AOPs) provide researchers and regulators a formal framework to link molecular initiating events (MIEs) to adverse outcomes (AOs) by defining causal relationships between measurable key events (KEs) and scientifically defensible underpinnings that support extrapolation from one even to the next the pathway. Because stressors will often trigger more than one pathway, and those pathways can interact and influence one another, it is important to consider AOPs in a systems context. AOP networks, which are single AOPs joined by one or more key events, can be developed and analyzed to provide a systems-level understanding. This paper highlights graph theoretic and network analytic techniques that are relevant for studying properties of AOP networks. These techniques may be applied by program office partners to use AOPs in the evaluation of more complex, “real world”, exposure scenarios. Additionally, this paper looks at the repository of AOP knowledge within the AOPwiki (AOPwiki.org) applies a set of network analysis techniques to benchmark the current state of the AOPwiki. This will be useful for guiding development of new AOPs within the CSS 17.01 Adverse Outcome Pathway Discovery and Development Project and providing baseline metrics to help assess the overall growth and impact of the AOP knowledgebase in upcoming years.
