We outline the following two aims. First, we intend to develop novel mediation frameworks that discover proteomic and metabolomic factors that mediate the effect of genetic variation on complex traits. Although numerous links between genetic variation and phenotypes have been discovered through genome-wide association studies (GWAS) and similar study designs, the bulk of these signals implicate regulatory rather than coding variation. Through similar approaches to Mendelian Randomization, we plan to develop bespoke causal inference methods that identify the downstream proteomic and metabolomic mediators of disease-associated genetic variation. Secondly, we are also in interested in the genetic epidemiology of clonal hematopoiesis – a phenomenon common in elderly individuals, where a clonal expansion in blood driven by a driver mutation occurs. The whole-genome sequencing available in UKB (of blood samples) will enable these analyses. We plan to identify recurrent somatic mutations in the UK Biobank, and then perform GWAS using the somatic mutations as phenotype to identify the germline basis of these somatic mutations.
