The sample was genotyped using Illumina’s Human660W-Quad Array (Illumina, Inc., San Diego, CA) following standard protocol. Of the 561,490 non-intensity SNP markers on the array, 527,829 (94.0%) were retained after eliminating markers that had 1) call rates < 99%, 2) minor allele frequency < 1%, 3) significant deviation from Hardy-Weinberg equilibrium at p < 1e-7, 3) more than two Mendelian inconsistencies across families or more than one mismatch in duplicated samples, 4) an association with participant sex or batch at p < 1e-7, or 5) been identified by Illumina as a bad marker on the array. Samples were eliminated if they had 1) more than 5000 non-calls, 2) low Gen_Call scores, 3) extreme homozygosity or heterozygosity, or 4) failed to have familial relationships or sex confirmed, suggesting a sample mixup. Analysis here is based on the 527,829 autosomal SNPs remaining following these data cleaning steps. Additional details concerning quality control procedures are given by Miller et al. (2012).
