S4: Adverse Outcome Pathway networks

Generation of computationally predicted Adverse Outcome Pathway networks through integration of publicly available in vivo, in vitro, phenotype, and biological pathway data.
Noffisat Oki, Douglas Connect
PRESENTING AUTHOR: 

Noffisat Oki

INSTITUTION / COMPANY : 

Douglas Connect

POSITION: 

Bioinformatics Scientist

AUTHOR(S): 

Authors: Oki O. Noffisat1, Bell M. Shannon3, Rong-Lin Wang4, Mark Nelms1,2, and Edwards W. Stephen2

1. Oak Ridge Institute for Science and Education, 2. U.S. EPA, 3. ILS/Contractor Supporting NICEATM, RTP, NC, 4. U.S. EPA, Cincinnati

REFERENCES: 

ABSTRACT CONTENT / DETAILS: 

The Adverse Outcome Pathway (AOP) framework is becoming a widely used tool for organizing and summarizing the mechanistic information connecting molecular perturbations by environmental stressors with adverse ecological and human health effects.

However, the conventional process for assembly of these AOPs is time and resource intensive and has been a rate limiting step in their development. To accelerate the process, we developed computationally predicted AOPs (cpAOPs) by association mining of data from publicly available databases.

A cpAOP network of ~21,000 associations was established among 105 phenotypes from TG-Gates rat liver data, 994 REACTOME pathways, 688 High-throughput (HT) assays from ToxCast, and 194 chemicals. A second network of 128,536 associations was generated by connecting 255 genes representing ToxCast biological targets to 4,980 diseases from the Comparative Toxicogenomics Database (CTD) using either HT screening activity from ToxCast for 286 chemicals or CTD gene expression changes in response to 2,330 chemicals.

Each network was separately evaluated by manual extraction of disease-specific cpAOPs and comparison with expert curation of the relevant literature. Both networks were then merged with ~130,000 publicly available phenotypes from PhenomicDB covering several model species, to create a single resource for automated cpAOP extraction and prioritization via the use of probability and weighting metrics.  

This provides a more comprehensive hypothetical AOP list than is possible by expert evaluation alone. A review of the most likely putative AOPs given an outcome of interest can then be performed by experts.

The views expressed in this abstract are those of the authors and do not necessarily reflect the views or policies of the U.S. EPA