The COGA sample was genotyped using multiple arrays, details of which have been reported.28 Briefly, a set of 47 000 high-quality variants that were typed on multiple arrays and duplicate individuals was used for the initial phase of data alignment. The full set of variants was imputed for each array using the appropriate ancestry-matched 1000 Genomes Phase 3 reference panel. Imputed SNPs with INFO scores <0.30 or individual genotype probability scores <0.90 were excluded, as were SNPs that did not pass Hardy-Weinberg equilibrium (HWE P < 10−6), and SNPs with a minor allele frequency less than 0.05%. Details on the genotyping QC and imputation of the CATS sample are described in detail elsewhere25,29 and in the online supplementary material.
