Determining causality and consequence of expression quantitative trait loci

Abstract

Expression quantitative trait loci (eQTLs) are currently the most abundant and systematically-surveyed class of functional consequence for genetic variation. Recent genetic studies of gene expression have identified thousands of eQTLs in diverse tissue types for the majority of human genes. Application of this large eQTL catalog provides an important resource for understanding the molecular basis of common genetic diseases. However, only now has both the availability of individuals with full genomes and corresponding advances in functional genomics provided the opportunity to dissect eQTLs to identify causal regulatory variants. Resolving the properties of such causal regulatory variants is improving understanding of the molecular mechanisms that influence traits and guiding the development of new genome-scale approaches to variant interpretation. In this review, we provide an overview of current computational and experimental methods for identifying causal regulatory variants and predicting their phenotypic consequences.

Figure 1. Fine-mapping of a cis-eQTL for IFT52 using whole genome sequencing

(A) Shows association between markers and gene expression in a European population using a combination of microarrays and genetic markers typed by the OmniExpress (700k markers, genome-wide). The multiple-testing significance level is marked by a horizontal dashed line. Here, the top associated SNP (purple) is 3’ of IFT52. (B) Rerunning the cis-eQTL association using whole genome sequencing data (5M markers) identifies a new, more significantly associated variant at the transcription start site of IFT52. Furthermore, this variant is in weak LD (r² between 0.2 and 0.4, light blue) with multiple 3’ variants suggesting that the original top SNP detected in panel A was not in fact the causal variant but was associated due to its linkage with the causal variant now more likely located at the transcription start site. It is also important to note that the multiple-testing significance level has become more stringent when testing eQTL in whole genomes due to testing more markers such that variants near SGK2 which were significant in the OmniExpress analysis are no longer equally significant in the whole genome analysis.