Human Genome-Wide Association Studies for Insulin-Related Traits

Publicly available human genome-wide association study (GWAS) data sets for fasting glucose, fasting insulin, and IR (unadjusted and adjusted for BMI) were retrieved from large meta-analysis consortia including MAGIC (the Meta-Analyses of Glucose and Insulin-related Traits Consortium) (26) and the GENESIS (Genetics of Insulin Sensitivity) Consortium (27). After retrieving summary-level statistics for all single nucleotide polymorphisms (SNPs), we removed SNPs with a weak association (<80%). The remaining SNPs with high linkage disequilibrium (r² > 0.5) were filtered by using a previously described method (28). Linkage disequilibrium data for European ancestry was obtained from HapMap3 (29) and the 1000 Genomes Project (30). Comprehensive lists of human cis–expression quantitative trait loci (eQTL) from human adipose, liver, and muscle tissues were accessible from our Mergeomics web server (31). Cis-eQTL were defined as eQTL in which the associated SNP and gene pairs are within 1 MB of each other. Details of the eQTL data sets used in the study are listed in Supplementary Table 1. A total of 1,690 coexpression modules were constructed from adipose, liver, and muscle tissue samples generated from multiple human and mouse studies (Supplementary Table 1) by using the WGCNA package in R software (32).