A unified framework for cell-type-specific eQTLs prioritization by integrating bulk and scRNA-seq data

Abstract

Genome-wide association studies (GWASs) have identified numerous genetic variants associated with complex traits, yet the biological interpretation remains challenging, especially for variants in non-coding regions. Expression quantitative trait loci (eQTLs) studies have linked these variations to gene expression, aiding in identifying genes involved in disease mechanisms. Traditional eQTL analyses using bulk RNA sequencing (bulk RNA-seq) provide tissue-level insights but suffer from signal loss and distortion due to unaddressed cellular heterogeneity. Recently, single-cell RNA sequencing (scRNA-seq) has provided higher resolution enabling cell-type-specific eQTL (ct-eQTL) analyses. However, these studies are limited by their smaller sample sizes and technical constraints. In this paper, we present a novel statistical framework, IBSEP, which integrates bulk RNA-seq and scRNA-seq data for enhanced ct-eQTLs prioritization. Our method employs a Bayesian hierarchical model to combine summary statistics from both data types, overcoming the limitations while leveraging the advantages associated with each technique. Through extensive simulations and real-data analyses, including peripheral blood mononuclear cells and brain cortex datasets, IBSEP demonstrated superior performance in identifying ct-eQTLs compared to existing methods. Our approach unveils new transcriptional regulatory mechanisms specific to cell types, offering deeper insights into the genetic basis of complex diseases at a cellular resolution.