Despite a large sample size, comprehensive gene lists, and a calibrated statistical method, we did not find evidence that common variants in proximity to ∼1,000 known nuclear-encoded mitochondrial genes contribute to T2D susceptibility. Similarly, we found no indication of significant associations between variants near these genes and intermediate physiological phenotypes related to T2D. Simulations of MAGENTA performance suggest that if there is a genetic contribution it is small - probably no more than 2–4% of nuclear-encoded mitochondrial genes (∼20–40 genes) harbor common variants of modest effect (e.g. an odds ratio of ∼1.07 for allele frequency of 0.2–0.3 and sample size of 10,000 individuals) on T2D risk. This number may vary to some extent depending on the actual effect sizes and total number of causal genes for the disease (see Figure S6). As of the latest T2D meta-analysis used here (DIAGRAM+), three mitochondrial genes (IDE, C8orf38 (Entrez ID 137682), and ACADS (Entrez ID 35)) lie near validated T2D SNPs amongst other genes in the interval [20], but a causal connection for these genes with T2D has not yet been shown.
