of cis-genetic effect on trait. In practice, for each gene, all SNPs within 1Mb of the gene present in the GWAS study were selected, and Σs,s and Σe,s were computed in the reference panel (i.e. expression and SNP data). To account for finite sample size and instances where Σs,s is not invertible, we adjusted the diagonal of the matrix using a technique similar to ridge regression with λ=0.1 (as evaluated in Pasaniuc et al4). This regularization, as well as noise in the estimation of W, can translate to lower power for association but yield conservative imputed Z-statistics.
