Systems level analysis of non-model organisms: a tool for understanding environmental stress

Omics techniques are changing the focus of ecotoxicology. In addition to challenges resulting from large amounts of data, there are further difficulties for non-model species: from lack of annotation to limited number of additional databases for molecular interactions and functions. In this thesis, I demonstrate the use of systems biology to relate molecular measurements to physiological parameters in non-model species in the context of environmental stress. Firstly, I make dynamical data-driven model of how gene expression changes in relation to the nerve conductance in earthworm Eisenia fetida exposed to single chemicals in the laboratory. The model reveals that gene expression changes might reflect the recovery from nerve damage. Using a similar approach, I use blue mussel Mytilus edulis sampled from their natural environment to model their annual cycle, integrating 1H-NMR metabolite levels with physiological and environmental parameters. I challenge this model created from data from a reference site to see site-effects for mussels sampled from an industrial harbour. Finally, I use systems biology to relate changing chemical concentrations and traditional toxicity assays in an effluent remediation system to stickleback gene expression and morphology. I demonstrate that data-driven systems biology can help the interpretation of complex problems. Supplementary data for this thesis can be found on the University of Birmingham eData repository at: https://doi.org/10.25500/edata.bham.0000065