The REML twin-based model can be used for eQTL analysis by including SNP genotype (additive coding as copies of the minor allele) and computing the corresponding Wald statistic, in this manner properly handling covariates and twin correlation structure. This approach is computationally prohibitive for full eQTL analysis, so we used Matrix eQTL 88 to rapidly screen for local or distant eQTL relationships. To account for dependence, the full REML model was then applied to all transcript-SNP associations with nominal P < 10−5 (a liberal threshold for the ~3×1010 tests performed). Separate false discovery rate (FDR q-value) error control was performed for local and distant eQTLs. After FDR correction it was apparent that all significant results with true REML q < 0.10 had indeed been captured. Some of the eQTL findings are reported in terms of unique genes, i.e. the most significant transcript-SNP combination for each gene, and in such instances the full testing multiplicity was considered.
