For genome‐wide significant signals, replication was attempted for the top five SNPs as well as with all SNPs within the nominated genes (with ±50 kb flanking regions). For the replication analyses, models identical to those applied in the discovery sample were used. To account for multiple testing, we applied a modified Bonferroni correction based on the effective number of independent SNPs in the gene regions calculated using a formula proposed by Gao and colleagues (Gao et al. 2008; Hendricks et al. 2014).
