In this study we performed a comprehensive eQTL mapping to explore the downstream effects of SNPs on gene expression by analyzing genotype and expression data of 1,469 unrelated samples. In addition to a genome-wide analysis, we also performed a focused analysis for disease- and trait-associated SNPs and SNPs located within the HLA. We replicated the identified trans-eQTLs in a collection of monocyte expression data and expression data from subcutaneous adipose, visceral adipose, muscle and liver tissue. Principal component analysis (PCA) enabled us to remove non-genetic expression variation [19], [20], resulting in increased power to detect eQTLs. A stringent probe-mapping strategy was used to filter out false-positive cis-eQTLs due to primer-polymorphisms and false-positive trans-eQTLs due to cross-hybridizations. Furthermore, a permutation strategy was utilized that corrects for multiple-testing, while preventing potential confounders such as non-even distribution of SNP markers and expression probe markers across the genome, differences in minor allele frequency (MAF) between SNPs, linkage disequilibrium (LD) within the genotype data, and correlation between expression probes.
