Novel statistical methodology for rare variant association in a case-control cohort (360G-Wellcome-203834_Z_16_A)

Genetic association studies focusing on common variation have uncovered only a fraction of proposed trait heritability. Some of this so-called missing heritability will be found within rare variation in the population. This hypothesis is supported by the facts that recent explosive population growth has increased the population burden of rare variants and deleterious variants are kept at low allele frequencies. All genetic susceptibility to disease is caused by alterations to the genes or their expression and for this reason it seems fruitful to focus an association study on the genes themselves. Any associations found are then directly informative about the molecular basis of disease without the need for fine mapping. The proposed project aims to develop a statistical method to find genes associated with disease by analysing the rare variation present in a case-control cohort. We aim to extend existing methods by including a previously unconsidered parameter; the position of the variants in a gene. In scenarios where differences in clustering or distribution of variants are observed between cases and controls, this method will have a substantial increase in power. This technique will be useful for elucidating the molecular mechanisms causing the disease and thus discovering new therapeutic targets.