Exploring the genetic architecture and the chromatin organisation of breast cancer

With breast cancer being a highly prevalent complex disease that affects many women worldwide, research over the years has focused on establishing underlying breast cancer risk factors. Understanding how, and why the disease develops will potentially reduce the number of women developing breast cancer, or increase the number of women being diagnosed at an earlier stage of development. The disease has been shown to be a highly polygenic trait, so in order to learn more about the disease, this thesis focuses on the polygenic basis of breast cancer. Two breast cancer GWAS, the UK2 and BBCS, and the COGS were used to conduct the analyses presented in this thesis. Using current chip heritability estimation methods, it was estimated that just under half of the genetic variation explained on the liability scale could be explained by genotyped SNPs. Common SNPs (MAF > 0.1) were shown to explain a substantial proportion of this variation, and the variance explained by each chromosome was shown to be linearly related to chromosome length, which indicated that variation is spread evenly across the genome. With BMI and age at menarche shown to be breast cancer risk factors, it was examined whether a shared polygenic basis exists between breast cancer and BMI, and whether there was evidence to suggest that breast cancer polygenic scores interact with either BMI, age at menarche or individual SNPs, to have an effect on breast cancer risk. With many susceptibility loci mapping to non-protein-coding regions of the genome, it was also tested whether individual genome-wide significant loci interact with other regions of the genome to influence breast cancer risk. These results give further insight into the polygenic architecture of breast cancer, and provide further evidence that a large number of genetic variants explain much of the genetic variation in breast cancer.