Genotyping was performed at the Wellcome Trust Sanger Institute using the Human670‐QuadCustom Illumina BeadChip (N = 1104) and the Illumina Human Core Exome BeadChip (N = 858). Pre‐imputation exclusion criteria for the data generated with the Human670‐QuadCustom Illumina BeadChip were minor allele frequency (MAF) < 0.01, sample and SNP call rate <0.95 (<0.99 for SNPs with MAF < 0.05); and the criteria for the data generated with the Illumina Human Core Exome BeadChip were minor allele count <2, sample call rate <0.98, SNP call rate <0.95 (<0.99 for SNPs with MAF < 0.05). Both genotype datasets were filtered according to the Hardy–Weinberg equilibrium (HWE) test P < 1e‐06. Further, sample heterozygosity test, gender, and Multidimensional Scaling (MDS) outlier checks were done for both. Pre‐phasing of the data was done with SHAPEIT2 (Delaneau et al. 2013) and imputation with IMPUTE2 (Howie et al. 2009) using the 1000 Genomes Phase I integrated haplotypes (produced using SHAPEIT2) reference panel (1000 Genomes Project Consortium, 2012). Quality controls and imputation for the GWAS data were done centrally at the Institute for Molecular Medicine Finland (FIMM), University of Helsinki, Helsinki, Finland.
