Abstract: Mode-of-action ontologies are gaining momentum in the field of toxicology and risk assessment. They are defined as frameworks qualitatively and quantitatively integrating and structuring relevant biological, toxicological, chemical and kinetic data from various sources. Ontologies have their roots in adverse outcome pathway networks, which in turn originate from physiological maps. Among other applications, ontologies can serve as the conceptual basis for setting up animal-free and human-relevant batteries for the toxicity testing of chemicals. This will be demonstrated in this presentation. Focus will be put on the liver, which is a frequent target for systemic toxicity because of its unique location and function in the organism. A tiered ontology-driven approach for the prediction of steatotic and cholestatic liver toxicity induced by chemicals and relying on combined in silico/in vitro testing as well as on expression and functional analysis will be presented.