out of 8 replication attempts only one was successful (for PTPRD, rs17584499) at P<0.05, when a total of 7.49 replications would be expected by power alone (4 for PTPRD and 3.49 for SRR, see Table S5). Also, in only 4 out of 8 cases the risk allele was the same (two for each gene). Overall, the replication attempts gathered in our database suggest that both associations were false positive findings in East Asians. These results make it clear that Fu et al. [21] were right in asking for caution, since putative population-specific associations may well turn out to be false positives. Moreover, the inclusion of more recent studies in our dataset helps discarding the population-specific status of some true associations. For instance, the association of 10p13 (CAMK1D, rs12779790) to type 2 diabetes was considered as European-specific, but it has been eventually replicated in East Asians [33].
