Aims

To develop new methodology for epidemiological datasets, with a particular emphasis on handling massive amount of sequence data. Specifically, we will detect and account for genetic structure in the form of known familial relationships, distant relatives, and population stratification. We will develop methodology to correctly use rare variant information in Association studies to determine the causes of disease from such datasets. Further, we will improve the use of genetic instruments in mendelian randomization studies by extending the developed methods to an instrumental variable context.