Beyond the biological interest of eQTL sharing among tissues, the availability of multiple tissues can also increase the power to detect eQTLs active in multiple tissues by combining information across tissues (24). To investigate, we conducted a permutation analysis by holding the expression data fixed and permuting the genotypes. In this manner, we identified the significant eGenes per tissue (as in the single-tissue analysis above, but for a smaller window near the TSS), or jointly for any combination of tissues by considering the minimum P value across tissues under each permutation. With this approach, the number of eGenes identified for individual tissues was similar to the single-tissue eQTL analysis (Fig. 2A). Next, the minimum P value for each gene across all nine tissues was used to test for eQTL evidence, and subjected to gene-level false discovery control. A total of 7425 eGenes with FDR < 0.05 were identified, representing a factor of 3 increase relative to the maximum number of significant eGenes for any single tissue. The Bayesian models, which leverage the high proportion of tissue-common eQTLs, increase the power
