We used the Genome-wide Complex Trait Analysis (GCTA) software (Yang et al. 2011) to estimate the amount of variance in initiation explained collectively by the SNPs. The aim of this analysis is to obtain an indication of the total signal in the SNPs, without identifying individual SNPs. Genetic similarity among the phenotyped individuals was computed based on best guess genotypes at 5 928 887 loci observed or imputed using the GoNL reference panel. The analyzed SNPs had a MAF larger than 1 %, imputation quality greater than 0.8 and showed no significant deviation from Hardy–Weinberg equilibrium given α = 1 × 10−4. The sample with observed initiation status (N = 6744 related individuals of Dutch ancestry) and the relevant covariates included in the genomewide SNP-based analysis (see below) were also used in the GCTA analysis. Furthermore, one of a pair of closely genetically related individuals (i.e., with an estimated genetic relatedness larger than 0.025) was dropped, which left for the analysis 3616 distantly related individuals. We specified the prevalence as equal to 22 %, value chosen in line with the
