Quantifying the role of linkage disequilibrium in mapping genetic regions within and across traits.

During the past decade, large-scale genetic studies have successfully identified tens of thousands of genetic variants associated to common diseases and have produced large repositories of summary association data across millions of individuals. This has provided great opportunities for subsequent analyses (e.g., to find unexpected relations among human traits) and has motivated the development of a wide-range of computational/statistical methodologies that leverage these large repositories of summary association statistics. A major component of all such methods that re-analyze summary association data is the incorporation of genetic variation data from small population-based catalogues of human genetic variation such as the 1000 Genomes. The discrepancy in the size of population-based catalogues (thousands of individuals) versus millions in disease studies induce biases that can yield false positive inference. Here we propose to leverage the large sample size of the UKBiobank to increase accuracy of summary association-based methods. We will use the newly developed methods to re-analyze real phenotypes from UKBiobank to identify new putative risk genes, to identify new risk regions and to identify new relations between traits and their subtypes.