Functional Architectures of Local and Distal Regulation of Gene Expression in Multiple Human Tissues

Abstract

Genetic variants that modulate gene expression levels play an important role in the etiology of human diseases and complex traits. Although large-scale eQTL mapping studies routinely identify many local eQTLs, the molecular mechanisms by which genetic variants regulate expression remain unclear, particularly for distal eQTLs, which these studies are not well powered to detect. Here, we leveraged all variants (not just those that pass stringent significance thresholds) to analyze the functional architecture of local and distal regulation of gene expression in 15 human tissues by employing an extension of stratified LD-score regression that produces robust results in simulations. The top enriched functional categories in local regulation of peripheral-blood gene expression included coding regions (11.41×), conserved regions (4.67×), and four histone marks (p < 5 × 10^-5 for all enrichments); local enrichments were similar across the 15 tissues. We also observed substantial enrichments for distal regulation of peripheral-blood gene expression: coding regions (4.47×), conserved regions (4.51×), and two histone marks (p < 3 × 10^-7 for all enrichments). Analyses of the genetic correlation of gene expression across tissues confirmed that local regulation of gene expression is largely shared across tissues but that distal regulation is highly tissue specific. Our results elucidate the functional components of the genetic architecture of local and distal regulation of gene expression.

Keywords: eQTLs; functional annotation; gene expression regulation; heritability.

Figure 1

Simulations Assessing Type I Error and Bias of Local Enrichment Estimates (A) Null simulations demonstrate well-calibrated type I error, given that estimated functional enrichments (average across ten simulations) are not statistically different from 1 after Bonferroni correction. Error bars represent 95% confidence intervals based on empirical SEs of the average enrichment across ten simulations. (B) Causal simulations demonstrate unbiased estimates of functional enrichments. Red dots represent the true expected enrichments. Center bars represent estimated enrichments (average across ten simulations). Error bars represent 95% confidence intervals based on empirical SEs of the average enrichment across ten simulations. Although some estimated enrichments lie just outside the 95% confidence intervals, they are not statistically different from the true enrichments after Bonferroni correction. Results are displayed for a representative set of 14 categories; numerical results for all 57 categories (for null and causal simulations) are reported in Tables S4 and S5.

Figure 2

Functional Enrichments for Local Regulation of Gene Expression (A) Local enrichment of each category in peripheral blood (Wright et al. dataset, n = 3,754). Error bars represent 95% confidence intervals. Results for the AUC metric are displayed in Figure S6. (B) Local enrichment of each category across 15 tissues. Purple shading indicates enriched categories (enrichment > 1), orange shading indicates depleted categories (enrichment < 1), and asterisks indicate significant enrichment or depletion after correction for 57 hypotheses tested. Sample sizes of each dataset are reported in Table 1. Results are displayed for a representative set of 14 categories; numerical results for all 57 categories are reported in Table S6, Tables S7 and S8. Estimates of the total local $h_{\text{g}}^2$ for each tissue (average across all genes) are provided in Table S18. A description of each functional category is provided in Table S2.

Figure 3

Comparison of Functional Enrichments for Local Regulation of Gene Expression in Peripheral Blood and Nine Complex Traits Enrichments for complex traits are meta-analyzed enrichments of nine complex traits and diseases from Finucane et al. Error bars represent 95% confidence intervals. The red dashed line represents y = x. Results are displayed for a representative set of 14 categories; numerical results for all 53 categories are reported in Table S9. H3K27ac (PGC2) is denoted as H3K27ac in the figure. A description of each functional category is provided in Table S2.

Figure 4

Functional Enrichments for Distal Regulation of Gene Expression (A) The distal enrichment of each category in peripheral blood (Wright et al. dataset, n = 3,754). Error bars represent 95% confidence intervals. Results for the area under curve (AUC) metric are displayed in Figure S6. (B) Comparison of functional enrichments for distal gene expression regulation versus local gene expression regulation in peripheral blood. Error bars represent 95% confidence intervals. The red dashed line represents y = x. H3K27ac (PGC2) is denoted as H3K27ac in the figure. A description of each functional category is provided in Table S2.

Figure 5

Pairwise Local and Distal Genetic Correlation across Tissues (A) Pairwise local genetic correlations across 11 GTEx tissues. (B) Local (upper left) and distal (lower right) genetic correlations across three MuTHER tissues. Numerical results are reported in Tables S16 and S17.