> "All of our analyses controlled for graduate/sibling status, age, gender, and the interactions of these variables, as well as the first three principal compon

> "All of our analyses controlled for graduate/sibling status, age, gender, and the interactions of these variables, as well as the first three principal components of the genetic data from the full set of 90 genotyped SNPs (to account for possible population stratification)." It is customary to do genomic control using likely-uncorrelated SNPs (i.e., excluding the 13 SNPs nominally reported to be in LD with the trait of interest). That they included these 13 when estimating pop strat from a small overall number of alleles makes me concerned that they may have unintentionally "controlled for" the actual effect, if any did exist. Even then, since they controlled for the first 3 PCs of that data, and I would assume the 13 candidate SNPs to be in linkage equilibrium, they shouldn't be getting rid of more than a couple actual associations, yet they find none replicate. So my observation is probably just a curio.