To Örebro University

Risk assessment plays a crucial role in evaluating and managing the potential hazards and health effects associated with exposure to substances. In recent decades, the field of risk assessment has undergone significant expansion, embracing innovative approaches and methodologies. Despite these advancements, it's noteworthy that the standards governing experimental setups have remained largely unchanged. This thesis focuses on enhancing environmental risk assessment methodologies, particularly in the context of exposure protocols, by incorporating toxicogenomics and machine learning approaches as well as suggestions for improved toxicity testing setup. The main objectives of the Paper I was to assess the sensitivity differences and shared responses of different animal models to exposure settings. Seven different organisms were tested with varying metal concentrations. Paper II investigated the effects of altering exposure media parameters, particularly water hardness. Paper III utilized computational advancements in toxicogenomics for gene ranking and exposure prediction. Paper IV investigated a larger number of genes by utilizing transcriptomics to discover novel biomarkers and molecular functions affected by metal exposures at the boundaries of Zn and Cu homeostasis. The research findings revealed that traditional toxicity assessment setups may not fully provide a base to capture the complexity of occurring toxicity. Therefore, the study proposes deviations from standard test protocols and emphasizes the need for holistic models that consider multiple factors to accurately assess toxicity risks in aquatic environments.