Longitudinal multi-dimensional investigation of metabolic and endocrine genetics

Thesis

Genome-wide association studies (GWASs) in recent decades have revealed the genetic landscape and shared aetiology of common, complex traits across the spectrum of human phenotypes. In this work, I develop and apply statistical tools to interrogate the genetic basis of, and relationships between, metabolic and endocrine traits. I demonstrate that under-explored primary care electronic health records (EHRs), linked to massive biobank projects across the globe, are a valuable source of longitudinal and rare biomarker data for genetics studies. Using EHRs, I find a common missense variant in the APOE gene that is associated with weight-loss in adulthood, which replicates in three global biobanking cohorts of between 125,000 to 475,000 individuals each. While the heritability of weight-change is low (< 3%), it has a distinct genetic architecture to baseline weight (genetic correlation, rG = 39%). Next, I perform GWASs for female and male infertility, as well as sex-specific reproductive hormones (with > 700,000 participants across seven global biobanks), to characterise the genetic contributions to these common but poorly understood phenotypes. I find 21 unique genetic loci for infertility, of which only six colocalise with reproductive hormone levels. While there is modest correlation between female infertility and heritable diseases of the reproductive tract, such as endometriosis (rG = 58%) and polycystic ovary syndrome (PCOS) (rG = 40%), I find no evidence for metabolic conditions such as obesity in the genetic aetiology of infertility. I explore these findings further through Mendelian Randomisation analyses to reveal heterogeneity in the genetically predicted causal effects of overall and central obesity on the risk of female reproductive conditions, including infertility, endometriosis, and PCOS, which may be partly genetically mediated by hormone levels. Through a range of genetics-based investigations, I outline the shared and distinct mechanisms of metabolic and endocrine disease in humans.

Authors: Sudheesh Venkatesh, S

• DOI: 10.5287/ora-14q5wwpaz
• Type of award: DPhil
• Level of award: Doctoral
• Awarding institution: University of Oxford
• Language: English
• Keywords: human genetics; infertility; women's health; reproductive health; statistical genetics; electronic health records; mendelian randomisation; genome-wide association study; sex hormones; longitudinal; obesity
• Subjects: Genetics; Obesity; Infertility; Reproductive health