Current research suggests that hundreds to thousands of single nucleotide polymorphisms (SNPs) with modest effect sizes contribute to the genetic basis of many disorders, a phenomenon labeled as polygenicity. Additionally, many such disorders demonstrate polygenic overlap, in which risk alleles are shared at associated genetic loci. However, there are currently no well-developed statistical methods that can be utilized to detect specific subsets of SNPs involved in the shared polygenicity of phenotypes. In this paper, we illustrate how elastic nets, with appropriate adaptation in selecting the penalty parameter, can be utilized for narrowing the range of SNPs involved in shared polygenicity. We first develop the method when individual-level data from genomewide association studies (GWASs) are available; we also extend the approach so that it can be used when only summary level data from GWASs are available. We illustrate and assess the performance of the proposed methods using extensive simulations, and by applying the methods to summary level data from a pair of related GWASs with fasting glucose level and BMI as the phenotypes.
School of Medicine