data (nominal p-value) was lower than the threshold of the 0.01 tail of the distribution of the minimal p-values (among all comparisons for a given gene) from 10,000 permutations of the expression phenotypes [30]. We calculated the false discovery rate for associations in each population at each threshold on the basis of the number of genes tested and the significance threshold of the permutations. We also estimated FDR by evaluating the degree to which associations that are discovered in one population replicate in another population. For each population, we considered all significant associations (0.01 permutation threshold) and determined whether the SNP-probe pair corresponding to the most significant SNP/Ensembl gene replicated (0.005 nominal and same direction) in at least one other population. We estimated FDR as 1−(the number of genes with replication/total number of significant genes).
