The role of genetic variation in type 2 diabetes predisposition

Genome wide association studies (GWAS) and next generation sequencing technologies (NGS) are proving to be transformative in facilitating our understanding of the genetic component of type 2 diabetes (T2D). These methodologies have identified common, low frequency and rare genetic variants that are associated with T2D predisposition and can provide valuable insight into disease pathophysiology. The work presented in this thesis uses a combination of approaches to understand how common and rare variants influence disease susceptibility. The initial waves of studies seeking common variant associations with T2D focused on individuals of European ancestry. Evaluating their results in other populations may offer insight into ethnic differences in disease prevalence and presentation as well as identify novel mechanisms for disease susceptibility. My work aimed to determine whether a subset of common variants conferring T2D-risk in Europeans contribute similarly to T2D susceptibility in South Asians, who have a substantially higher occurrence of T2D. My analysis identified no systematic difference in allelic odds ratios or allele frequencies between samples of Sri Lankan and European origin. This suggests that systematic differences in common variant associations are unlikely to explain the interethnic differences in prevalence and presentation of T2D. As part of a large-scale collaborative GWAS in individuals of South Asian ancestry, my work contributed to the identification seven loci revealing novel mechanisms, which may contribute to diabetes risk. Genome-wide association study results have also indicated several biomarker profiles (glycan profiles and hsCRP) that might be useful diagnostic tools for discriminating between patients with Maturity-Onset Diabetes of the Young (MODY) caused by mutations in HNF1A (HNF1A-MODY) and common forms of diabetes. Correctly, diagnosing HNF1A-MODY has important implications for patients' treatment and disease prognosis, and for their family members. However, cases of HNF1A-MODY are currently often incorrectly diagnosed. Biomarkers may therefore offer a useful way to prioritise patients for genetic screening. To examine whether glycan profiles and hsCRP could serve as surrogate measures for HNF1A dysfunction, I completed capillary sequencing of HNF1A in patients with biomarker profiles suggestive of HNF1A dysfunction. Using an evaluation pipeline comprising family studies, in silico evaluation and in vitro functional characterisation, I demonstrated that variants within HNF1A are associated with a spectrum of phenotypes varying from casual for MODY, increasing T2D risk or neutral. My work exemplifies the considerable challenges in determining the clinical significance of coding variants identified through NGS on disease risk. The data shows that biomarker profiles are a promising tool for assessing HNF1A dysfunction. However, it is likely that clinical interpretation of variants within HNF1A will need to be performed on an individual basis using a combination of factors that includes clinical, biochemical and lifestyle factors.