Analysis of Genetic Variation in Humans and Other Species

Recent advances in sequencing technologies have led to the generation of vast amounts genetic variation data for many species, including humans, with advances in our understanding of disease now limited by the speed at which these data can be analysed. This thesis focuses on the analysis of genetic variation at multiple levels. First, in the human disease cystinuria, an inherited form of kidney stones. Genetic variants previously associated with cystinuria were characterised using a series of computational methods, identifying key functional features of these mutations. Predictions of disease severity for a cohort of 74 cystinuria patients were then made based on the genotypes of each individual. When compared to clinical outcomes, these predictions demonstrate the potential for computational methods in delivering precision medicine to cystinuria patients. Second, a genome-wide analysis of variant combinations in individual human genomes identified combinations of variants protein-wide, within close spatial proximity in the 3-dimensional structures of proteins, and in protein-protein interface sites. The vast majority of computational methods for analysing genetic variation consider only one variant at a time. This work highlights the importance of analysing the combined effects of variants, which will be a key challenge in the future of computational biology and precision medicine. Finally, two different analyses of ebolaviruses were performed. The first study focused on human pathogenicity of ebolaviruses, a critical challenge in epidemiology. This study identified a set of key variants that differentiate human pathogenic and non-pathogenic ebolaviruses. The second study focused on the evolution of the Ebola virus genome, the most common causative species of human ebolavirus outbreaks. Ebola virus genome evolution was analysed over time since its identification in 1976, and over the course of the 2013-2016 West Africa outbreak. A strong bias for transition mutations was identified, with suspected mutational pressure from host APOBEC and ADAR enzymes.