In order to confirm our possible novel menarche loci, we tested our 42 top hits for in silico association with age at menarche in 8,669 women from 16 studies with GWAS data, and which were not included in the first stage meta-analysis (Supplementary Table 4). In addition, new genotype data was generated for 30 of the 42 menarche loci and tested for association with age at menarche in up to 6,118 women from the Avon Longitudinal Study of Parents and Children (ALSPAC). Genotyping was performed by KBiosciences (Hoddesdon, UK) using their own novel system of fluorescence-based competitive allele-specific PCR (KASPar). As in stage 1, analyses were restricted to women reporting age at menarche between 9 and 17 years, and adjustment was made for birth year. Mean age at menarche ranged from 12.4 to 13.5 years, consistent with studies in the stage 1 meta-analysis. Linear regression was used to test the association between each variant and age at menarche in an additive genetic model. These results were then meta-analysed with GC-adjusted statistics from our stage 1 meta-analysis using inverse-variance fixed effects models.
