The goal of the proposed work is to develop computational and statistical methods for analyzing large-scale genetic and phenotypic data. These methods include fast methods for computing genetic structure and ancestry, for associating genetic variants with phenotypes, risk prediction, testing associations and for imputing genotype and phenotype. The UK Biobank data will be used to develop and benchmark these methods. The proposed methods will lead to better understanding of the genetic basis of diseases and will lead to improved methods for prediction and treatment. We plan to use the rich UK Biobank data to develop scalable methods for analyzing genetic data and for finding genetic variants associated with phenotype. We would like access to the full cohort (genetic and phenotypic data).