FastGxC: Fast and powerful context-specific eQTL mapping in bulk and single-cell data

Context-specific expression quantitative trait loci (eQTLs) mediate genetic risk for complex diseases, but current methods limit their characterization and interpretation. We introduce FastGxC, a method for efficiently mapping context-specific eQTLs by leveraging correlation structure in multi-tissue bulk and single-cell RNA sequencing studies. In simulations, FastGxC is nine times more powerful and 10⁶ times faster than existing approaches, reducing computation time from years to minutes. We applied FastGxC to bulk multi-tissue (n = 698) and peripheral blood mononuclear cell (PBMC) single-cell RNA sequencing datasets (n = 1,218), generating comprehensive tissue- and cell-type-specific eQTL maps. These eQTLs showed 4-fold enrichment in context-matched open chromatin and were twice as enriched as standard context-specific eQTLs. FastGxC improved precision in identifying relevant trait contexts by 3-fold and expanded candidate causal genes by 25% in cell types and by 6% in tissues. FastGxC provides a powerful framework for mapping context-specific eQTLs, advancing our understanding of gene regulatory mechanisms underlying complex traits.