To obtain a more precise assessment of the potential size of reporting bias in East Asian GWAS, we estimated the maximum number of failed (P>0.05) but unreported replication attempts that is actually possible in our database [30]. Specifically, and for each East Asian GWAS, we computed the number of disease-associated SNPs discovered in Europeans that were not mentioned in East Asian studies (neither a p-value nor any other information was reported in the main text or in the supplemental information). In total, we found 416 such instances. Most of these cases may not constitute reporting bias at all, since the SNPs in question may not be included in the array used for the East Asian GWAS, may be monomorphic in the studied population, may have been filtered out during QC and so on. Still, making the extreme assumption that all these 416 cases are failed replications, they constitute the maximum number of biased reports that we could have not included in our database (Table S8). Under this extreme scenario, the 103 positive replications would not have been drawn from the
