We performed the single‐variant association analyses only for those SNPs with a MAF > 5% and HWE test P > 1e‐05, and ensured that identified top signals had high imputation information score (>0.8). The total number of SNPs included in the genome‐wide time‐to‐event analyses was 5,918,992. We checked for potential population stratification by investigating the principal components for the family founders with the EIGENSOFT package (Price et al. 2006); no outliers with foreign ancestry were found, as was expected as the twin data consist purely of native Finnish population. We adopted the genome‐wide significance P‐value of P < 5e‐08 as a cutoff, which has been broadly recognized as a criterion based on the sequencing data of European populations (Sham and Purcell 2014). For chromosomes containing loci exceeding the cutoff, we performed conditional analyses where we adjusted for the top SNP to test whether the detected association represented an independent signal. We list top five SNPs, regardless of their P‐values, to allow for comparison with results from the follow‐up studies adjusting for the relevant covariates. For SNPs identified as significantly associated
