Local genetic effects on gene expression across 44 human tissues

Abstract

64Expression quantitative trait locus (eQTL) mapping provides a powerful means to identify func-65 tional variants influencing gene expression and disease pathogenesis. We report the identification 66 of cis-eQTLs from 7,051 post-mortem samples representing 44 tissues and 449 individuals as part 67 of the Genotype-Tissue Expression (GTEx) project. We find a cis-eQTL for 88% of all annotated 68 protein-coding genes, with one-third having multiple independent effects. We identify numerous 69 tissue-specific cis-eQTLs, highlighting the unique functional impact of regulatory variation in di-70 verse tissues. By integrating large-scale functional genomics data and state-of-the-art fine-mapping 71 algorithms, we identify multiple features predictive of tissue-specific and shared regulatory effects. 72 We improve estimates of cis-eQTL sharing and effect sizes using allele specific expression across tis-73 sues. Finally, we demonstrate the utility of this large compendium of cis-eQTLs for understanding 74 the tissue-specific etiology of complex traits, including coronary artery disease. The GTEx project 75 provides an exceptional resource that has improved our understanding of gene regulation across 76 tissues and the role of regulatory variation in human genetic diseases. 77 Introduction 78 Genome-wide association studies (GWAS) have identified a wealth of genetic variants associated 79 with complex traits and disease risk. However, characterizing the molecular and cellular mechanisms 80 through which these variants act remains a major challenge that limits our understanding of disease 81 pathogenesis and the development of therapeutic interventions. Expression quantitative trait locus 82 (eQTL) studies provide a systematic approach to characterize the molecular consequences of genetic 83 variation across tissues and cell types 1-4 . Multiple studies have identified eQTLs for thousands of 84 genes 5-7 , providing novel insights into gene regulation and enabling the interpretation of GWAS 85 signals 8-12 . These studies have largely been performed in a few easily accessible cell types and cell 86 lines, precluding interpretation of the systemic and tissue-specific consequences of genetic variation. 87To overcome these limitations, the Genotype Tissue Expression (GTEx) project was designed to 88 identify and characterize eQTLs across a broad range of tissues. During the pilot phase, which 89 focused on nine tissues, the GTEx project highlighted patterns of eQTL tissue-specificity and 90 demonstrated the value of multi-tissue study designs for identifying causal genes and tissues for 91 trait-associated variants 1 . These results indicated that the identification of eQTLs across an even 92 broader range of tissues would drastically improve characterization of the gene-and tissue-specific 93 consequences of genetic variants. 94Here, we report on the discovery of cis-eQTLs across an expanded collection of 44 tissues in 95 the GTEx V6p study. This dataset consists of 7,051 transcriptomes from 449 individuals and 96 4...

Genetic regulatory effects modified by immune activation contribute to autoimmune disease associations

“…8b). While reQTL enrichment has been previously described for some transcription factors 5,6 , and annotations of condition-specific epigenomic marks and tissue-specific eQTLs have been described 8,9 , our results are to our knowledge the first demonstration of environmentally responsive eQTLs being enriched in distal cis-regulatory elements.…”

“…These results indicate a substantial, disease-specific role of environmental interactions with microbial ligands in genetic risk to complex autoimmune diseases. While tissue-specificity of molecular effects of GWAS variants is increasingly appreciated and analyzed 9 , our results suggest that innate immune stimulation is a key cellular state to consider in future eQTL studies as well as in targeted functional follow-up of GWAS loci.…”

A complete tool set for molecular QTL discovery and analysis

“…7). The running times are actually so small that it becomes possible to process rapidly massive data sets such as the GTEx v6p study 16 (7,051 samples in B870 CPU hours; Supplementary Fig. 8) and to repeat the whole analysis multiple times across different sets of quantifications, covariates and QC filters to determine the optimal configuration which maximizes the number of discoveries ( Supplementary Figs 9 and 10).…”

Genetic-Variation-Driven Gene-Expression Changes Highlight Genes with Important Functions for Kidney Disease

“…Next, we performed a multi-tissue eQTL meta-analysis using the METASOFT software. 51 For each gene, the proportion of tissues for which the gene had a strong eQTL effect was calculated (i.e., the proportion of tissues with m value R 0.9). And finally, tissue-specific eGenes were defined as genes in the bottom 10% of the empirical distribution of this proportion; for example, fewer than five tissues showed an m value R 0.9.…”