Development of adverse outcome pathway (AOP) of silver nanoparticles using comparative toxicogenomics database (CTD) and AOP wiki

다국어 초록 (Multilingual Abstract)

An adverse outcome pathway (AOP) is a framework that organizes the predictive relationships between molecular initiating events (MIEs), key events (KEs), and adverse outcomes (AOs), which contributes to providing evidence based mechanistic insights. Meanwhile, integrating and utilizing the massive and heterogeneous data from different biological levels is a laborious and time intensive process, brings a challenge in AOP’s construction. Fortunately, mining available publicly database through computational and bioinformatics approaches can make the mass of biological information into better serve as the basis for creating expert-derived AOPs in a much more efficient way. The Comparative Toxicogenomics Database (CTD, http://ctdbase.org/), as a comprehensive, continuously updated and publicly available database, provides manually curated associations between and among chemicals, genes, phenotypes, diseases captured from the scientific literatures. The chemical-phenotype module of CTD provides numerous opportunities for users to unravel the complex interaction between environmental exposure and molecular and functional changes in organisms. As a part of OECD AOP program, the AOP Wiki (http://aopkb.org) serves as the primary repository of qualitative information for the international AOP development effort, provides insights to identify the upstream or downstream of KEs. We present a case study discerning the role of silver nanoparticles (AgNPs) in influencing phenotypes preceding infertility, a disease related to reproductive toxicity. Combining CTD’s phenotype-disease inference data, AgNPs-phenotypes were anchored to infertility. Through integrating analysis in AOP wiki knowledge base, filtered phenotypes were linked to key events consisting of AOPs and assembled together based on upstream and downstream relationships. A subset of 13 phenotypes were filtered from CTD as potential key events. Selected phenotypes were used to construct a predictive AOP connecting AgNPs to infertility. Independent MIEs represent a curated interaction between AgNPs and 11 genes are directly annotated to infertility from CTD’s disease core. This study will help address the existing knowledge gaps for future experimental verification. Furthermore, this study will help discover and connect information to expand the predictive AOPs, as well as interrelate other AOPs to generate hypotheses about chemical exposures and environmentally influenced disease.