signals of association credited to our common SNPs might be synthetic associations resulting from the contributions of multiple rare SNPs in the IPO11-HTR1A region, which needs to be identified by sequencing. Third, the associations in the European-American discovery cohort, the associations in the replication cohorts, and the functional signals in the eQTL analysis did not perfectly match, which was probably because these risk markers were not the causal variants per se, but rather in LD with a common putative causal variant. Fourth, current evidence, including the effect sizes and the significance strength of associations, was not sufficient to fine-map the putative causal variant to any one of the four genome-wide significant risk markers, although the most significant one (i.e., rs7445832) was most likely. Sequencing is warranted to detect the actual causal variant. Finally, after conditioning on rs7445832, all association signals for other markers were significantly reduced, which might suggest there exists only one putative causal locus in this region.
