Advances in analytical chemistry have made it possible to discover many chemicals present in the environment that are potentially dangerous to humans and the environment. In order to protect people and the environment, we need to understand the biological targets of these chemicals and assess their harmful effects. Currently, the assessment of chemical toxicity is limited to the measurement of selected physiological endpoints on a limited selection of model organisms, which is not favourable due to the high number of chemicals requiring testing. A potential solution is to move to a more mechanistic approach, replacing testing at selected physiological endpoints with high-throughput molecular tests based on cell cultures, and interpreting their results using computerized knowledge sources, e.g. adverse outcome pathways (AOP). Over the last decade, the AOP concept has become one of the most powerful tools in environmental toxicology for assessing the toxicity of chemicals, but the development of AOP is still a slow and manually demanding process.The purpose of the master's thesis was to determine whether it is possible to develop new AOPs based on existing biological causal networks. We used neurotoxicological and cardiotoxicological biological causal networks developed at the National Institute of Biology for the zebrafish (Danio rerio) as the starting point for the development of the AOPs.In the initial phase of AOP development, the data set of biological causal networks of neurotoxic and cardiotoxic outcomes was manually curated in order to select the key network data needed to develop a network of potential AOPs. The selected data set was then visualized in the computer program Cytoscape, followed by network filtering, which removed unrelated network entities and pathways that did not lead to pathologies. Thus, we created a connected unit of the initial network, where all the entities were unique, called a network of potential AOPs. Such network was then broken down into smaller subnetworks. Subnetworks made it easier to review the linear paths involved in network of potential AOPs and highlight the most appropriate ones for development into new AOPs.This was followed by a manual review of the subnetworks and the selection of the most appropriate linear pathways for AOP development based on the selection criteria. Based on the OECD guidelines, two selected linear pathways were developed into AOPs, for which we described the building blocks, estimated the weight of the evidence of the AOP building blocks, and published them in the AOP-Wiki database.
|
