| Summary: | The aetiology of type 2 diabetes is complex with both genetic and environmental factors playing a role. Largely through genome-wide association studies, some proportion of the genetic component of type 2 diabetes and its related traits has been identified. However, there are some limitations in those study designs including possible overestimation of disease risk in the general population and a large proportion of unexplained heritability in the disease or trait of concern. There are several approaches that can be taken, to better understand the aetiology of type 2 diabetes. These include studying closely related quantitative traits such as fasting glucose levels and expanding the current efforts in discovering susceptibility variants to other classes of variation beyond common variants. Although, observational associations between type 2 diabetes and related quantitative traits may provide further clues to the aetiology of type 2 diabetes and its related traits causal nature of such associations are not clear. This thesis presents a series of studies that address the above concerns, and explore approaches of how known common variants associated with metabolic traits can be used to understand their role in the aetiology of the associated disease or trait. Chapter 1 is an introduction to the molecular genetics of type 2 diabetes, fasting glucose and birth weight and discusses the biological implications of these genetic findings. Chapter 2 describes a study that shows how genetic factors are important regardless of how patients are recruited using a common variant robustly 2 ociated with type 2 diabetes in the TCF7L2 locus. This was the first study to empirically show that the risk estimates from genetic studies involving genetically enriched cases are likely to be overestimated. Chapter 3 describes a study that was designed to identify rare or low frequency variants associated with fasting glucose levels in the G6PC2 gene where common variation in the same gene is robustly associated with fasting glucose levels. We found no evidence that the identified rare variants were associated with normal variation in fasting glucose levels. Chapter 4 describes a study that was designed to test the hypothesis that common genetic variants associated with fasting glucose and triglyceride levels would alter the same traits in pregnancy and influence fetal growth and development. We found evidence that the common variants alerting fasting glucose and triglyceride levels outside pregnancy largely alters these traits in pregnancy to the same extent. We also found preliminary evidence that the relationship between raised mothers glucose levels and increased birth weight is causal. Chapter 5 describes a study that utilises the Mendelian randomisation approach to test whether the observed correlations between raised circulating triglyceride levels and type 2 diabetes, fasting glucose and fasting insulin levels are causal. We found no evidence that raised circulating triglycerides are causal for type 2 diabetes, fasting glucose or fasting insulin levels. The observed correlations are therefore confounded or likely to be secondary to the diabetes disease process.
|
|---|
