|
There has been a dramatic increase in the variety and scale of genetics and genomics data over the past decade. It is common now to have a variety of omics measurements on the same subjects, from whole-genome sequence variation to epigenetic markers and microbiome. I will discuss some of methodological challenges in developing models and methods that combine and fully utilize these diverse data types.
I will use studies in the South Dakota Hutterites to illustrate methods for integrating whole genome sequencing, array SNP data, RNA-sequencing and pedigree information in gene mapping studies. I will describe advantages of using founder populations, including cost-efficient study designs. The efficiency is achieved using a fast phasing and imputation method that obtains accurate genotypes and high call rates in 1317 related Hutterites using whole genome sequencing data on only 98 related individuals. In addition, the algorithm allows accurate parent-of-origin assignments for each allele as well as imputed genotypes of recent ancestors with no available DNA. I will illustrate with applications to discovery of parent-of-origin effects in disease and eQTL studies.
|
