Among the several attempts to identify genes that explain the heritability of initiation, a linkage study (Agrawal et al. 2008a) failed to identify statistically significant associated genomic regions, although it did identify several suggestive regions on chromosomes 18 and 1. Likewise, a meta-analysis by Verweij et al. (Verweij et al. 2013) combining the results of two genomewide association studies (GWAS) comprising about 10 000 individuals failed to detect common single nucleotide polymorphisms (SNPs) associated with initiation. It should be noted, however, that the association analysis by Verweij and colleagues was limited to common (i.e., minor allele frequency (MAF) > 5 %) HapMap SNPs (Consortium 2010). With the recent completion of large sequencing projects such as the 1000 Genomes (1000G) (Consortium 2012) and the Genome of the Netherlands (Boomsma et al. 2014; The Genome of the Netherlands 2014), more detailed genotypic information has become available in large GWAS samples. Given the availability of denser SNPs, which are expected to be in high linkage disequilibrium (LD) with the causal variants, we aim to extend the search for genetic variants (GVs) implicated in
