SNP-to-gene linking strategies reveal contributions of enhancer-related and candidate master-regulator genes to autoimmune disease

Abstract

Gene regulation is known to play a fundamental role in human disease, but mechanisms of regulation vary greatly across genes. Here, we explore the contributions to disease of two types of genes: genes whose regulation is driven by enhancer regions as opposed to promoter regions (Enhancer-driven) and genes that regulate many other genes in trans (Master-regulator). We link these genes to SNPs using a comprehensive set of SNP-to-gene (S2G) strategies and apply stratified LD score regression to the resulting SNP annotations to draw three main conclusions about 11 autoimmune diseases and blood cell traits (average N =306K). First, Enhancer-driven genes defined in blood using functional genomics data (e.g. ATAC-seq, RNA-seq, PC-HiC) are uniquely informative for autoimmune disease heritability, after conditioning on a broad set of regulatory annotations from the baseline-LD model. Second, Master-regulator genes defined using trans-eQTL in blood are also uniquely informative for autoimmune disease heritability. Third, integrating Enhancer-driven and Master-regulator gene sets with protein-protein interaction (PPI) network information magnified their unique disease signal. The resulting PPI-enhancer gene score produced >2x stronger conditional signal (maximum standardized SNP annotation effect size (τ*) = 2.0 (s.e. 0.3) vs. 0.91 (s.e. 0.21)), and >2x stronger gene-level enrichment for approved autoimmune disease drug targets (5.3x vs. 2.1x), as compared to the recently proposed Enhancer Domain Score (EDS). In each case, using functionally informed S2G strategies to link genes to SNPs that may regulate them produced much stronger disease signals (4.1x-13x larger τ* values) than conventional window-based S2G strategies. We conclude that Enhancer-driven and Master-regulator genes are uniquely informative for human disease, and that PPI networks and S2G strategies magnify these signals.

A single-cell and spatial atlas of autopsy tissues reveals pathology and cellular targets of SARS-CoV-2

“…To relate COVID-19 severity traits to cell type level processes in each tissue, we analyzed gene programs characterizing cell types in general, as well as disease associated programs that are differentially expressed in the same cell type between COVID-19 and healthy tissue ( Methods ). We used sc-linker, a new computational approach ( Methods , KAJ, KKD, ALP, AR, unpublished work ), to link gene programs defined by scRNA-Seq to genetic signal using linkage disequilibrium (LD) score regression 61–63 .…”

“…We used sc-linker, a new computational approach (Methods, KAJ, KKD, ALP, AR, unpublished work), to link gene programs defined by scRNA-Seq to genetic signal using linkage disequilibrium (LD) score regression [61][62][63] .…”

Identifying disease-critical cell types and cellular processes across the human body by integration of single-cell profiles and human genetics

“…More fine-grained enhancer-gene linking strategies will likely prove beneficial, but the strategies that we used here provide a clear improvement over standard gene window-based approaches. We did not perform a comprehensive evaluation of enhancer-gene linking strategies and methods to combine them, which will be provided elsewhere( 155, 156 ) (S. Gazal, unpublished data). Second, we focus on genome-wide disease heritability (rather than a particular locus); however, our approach can be used to implicate specific genes and gene programs.…”

“…Here, we primarily considered an enhancer-gene linking strategy defined by the union of the Roadmap( 21, 172 ) and Activity-By-Contact (ABC)( 22, 23 ) strategies. Roadmap and ABC enhancer gene links are publicly available for a broad set of tissues and have been shown to outperform other enhancer-gene linking strategies in previous work( 155 ). We consider tissue-specific Roadmap and ABC enhancer-gene linking strategies for gene programs corresponding to any of the biosamples (cell types or tissues) associated with the relevant tissue.…”

Integrative approaches to improve the informativeness of deep learning models for human complex diseases

“…For comparison purposes, we also consider a simpler logistic regression model. Second, we improve the specificity of these annotations by restricting them to SNPs linked to genes using 10 (proximal and distal) SNP-to-gene (S2G) strategies [24][25][26][27][28][29][30][31][32]38 (Table 1). Third, we predict gene expression (and derive alleliceffect annotations) from deep learning annotations at SNPs implicated by S2G linking strategies, generalizing the previously proposed ExPecto approach 4 , which incorporates deep learning annotations based on distance to TSS.…”