In 2010, 175 authors published a paper in Nature Genetics (42:1077–1085) in which they used a meta-analysis of 32 genome-wide association studies in 87,802 women of European descent, with the aim of discovering “new loci for age at menarche.” They discovered thirty new such loci for age at menarche. By testing the 42 top hits for an in silico association with age at menarche, they managed to explain 1.3% of the variance in age at menarche. If one divides 1.3% by 175, then each author of the article contributed to explaining 0.0074% of the variance.
Four years later, 204 authors and 6 consortia made of 162 additional people have extended the same study and have published in Nature a paper on a meta-analysis of 57 genome-wide association studies in 132,989 women of European descent. In a further 49,427 women, data were available on up to approximately 25,000 single nucleotide polymorphisms (SNPs). The number of SNPs that reached the genome-wide significance threshold (P smaller than 0.00000005) for association with age at menarche was 3,915. By using a mysterious computational tool called GCTA, 123 independent signals for age at menarche at 106 genomic loci were identified. These signals included 11 loci containing multiple independent signals. The 123 SNPs together explained 2.7% of the variance, i.e., 1.4% more than previous study. Dividing 1.4% by 366, the estimate for the contribution to explaining the variance per author is 0.0038%.
There are much better uses for the money currently spend on GWAS