needed, and then SNP genotypes were tested for Mendelian inconsistencies (Pedcheck; O’Connell & Weeks, 1998) with the revised family structure. The cleaned genotype data were imputed to 1000 genomes (EUR and AFR, Phase 3, b37, October 2014) with build hg19 using SHAPEIT (https://mathgen.stats.ox.ac.uk/genetics_software/shapeit/shapeit.html) and IMPUTE2 (https://mathgen.stats.ox.ac.uk/impute/impute_v2.html). To avoid ambiguities in strand designation, SNPs with A/T or C/G alleles were removed. After imputation, genotype probabilities ≥ 0.90 were changed to genotypes. Mendelian errors in the imputed SNPs were reviewed and resolved as described in Wetherill et al., 2015 (Wetherill et al., 2015). All SNPs with imputation genotyping rate < 98% and MAF < 0.03 were excluded from association analyses.
