Protein-protein interactions shape trans-regulatory impact of genetic variation on protein expression and complex traits

Abstract

Most genetic variants influence complex traits by affecting gene regulation. Yet, despite comprehensive catalogs of molecular QTLs, linking trait-associated variants to biological functions remains difficult. In this study, we re-analyzed large maps of protein QTLs (pQTLs) to show that genes with trans-pQTLs but without cis-pQTLs are under strong selective constraints and are highly enriched in GWAS loci. We found that trans-pQTLs and their trans targets are highly enriched in interacting protein pairs, and trans-pQTLs in coding regions are significantly enriched at protein-protein interactions (PPI) interfaces. By leveraging existing PPI annotations for trans-pQTL mapping, we identified 26,028 trans-pQTLs influencing 1,061 PPI clusters. The trans-pQTLs of PPIs colocalized with 66% GWAS loci per trait on average for 50 complex traits, helping in many cases to link GWAS loci to cellular function. Finally, we identified trans-pQTL effects at multiple autoimmune GWAS loci that converge on the same PPIs, pinpointing protein complexes and signaling pathways that show promising therapeutic target potential.