One limitation of our eQTL analysis is that, like most eQTL analyses, it does not distinguish between causal associations and those that are due to LD. This issue is particularly important to appreciate when, for example, cross-referencing GWAS associations with eQTL effect estimates; a GWAS-associated SNP may be a “significant” eQTL simply because it is in LD with another causal SNP. For single-tissue eQTL mapping, this problem has been addressed in several ways, including fine-mapping24–29 and co-localization30–32. For multi-tissue analysis, only more limited attempts exist to address this problem. For example, eQTLBMA5 implements a Bayesian approach to fine-mapping under the simplifying assumption that there is at most one causal SNP per gene24,25. However, this assumption becomes less plausible in analyses of many tissues, and developing more flexible multi-tissue fine-mapping methods seems an important future direction.
