(ADVerse outcome Network for non-mutagenic CarcinogEns)

The Adverse Outcome Pathway (AOP) concept as a tool for gathering and linking of information at different levels of biological organization has been largely accepted by regulatory bodies and its usability has been recognized by scientists. The process of AOP generation, however, is still mostly done manually by experts screening through evidence and extracting probable associations.

The goal of the ADVaNCE project is to accelerate  this process and increase the reliability of the findings, therefore, we have developed an automated workflow for AOP hypothesis generation.

In brief, association mining methods were applied to high-throughput screening, gene expression, in vivo and disease data for chemicals was gathered from ToxCast and the Comparative Toxicogenomics Database (CTD), and subjected to  find relationships between genes, pathways and diseases that co-occur across datasets. This was supplemented by pathway mapping using Reactome to fill in gaps and identify events occurring at the cellular/tissue levels.  Furthermore, in vivo data from TG-Gates (using several time-points and dose levels) was integrated to finally derive a gene, pathway, biochemical/hematological, histopathological and disease information network from which specific disease sub-networks can be queried.

To test the workflow, non-genotoxic-induced hepatocellular carcinoma (HCC) was selected as a case study. The implementation of the workflow resulted in the identification of several non-genotoxic-specific HCC-connected biomarker genes belonging to two major categories, namely cell cycle proliferation and endoplasmic reticulum stress. Biochemical findings revealed non-genotoxic-specific alkaline phosphatase increase. On the other hand, the explored non-genotoxic-specific histopathology was mostly connected to pre-fibrotic to cirrhotic stage.

This workflow and case study further illustrates the utility of show computationally predicted constructs in supporting the process of AOP development by using  pre-existing knowledge in a fast and unbiased manner.