Abstract
The proteome is fundamental to human biology and health but little is known about ancestral diversity of its genetic determinants. In GWAS of plasma levels of 2,923 proteins in 3,974 Chinese adults, we identified pQTLs for 1,784 proteins, including 1,312 cis-pQTLs for 1,264 proteins. Fine-mapping identified 3,475 credible sets for independent pQTLs, 36% of which were distinct from those identified in European adults as assessed by three different methods. In phenome-wide MR analyses, 59 disease associations were in strong LD (r² >0.95) with a cis-pQTL sentinel variant, including 26 with strong evidence of colocalisation (PP.H4>0.9) and 34 not identified in previous similar European studies. Evaluation of current drug development identified 8 confirmed therapeutic targets, 8 potential repurposing targets, and 19 potential novel targets (13 not previously reported in Europeans). The findings demonstrate the importance of extending proteogenomic studies to non-European ancestry populations to identify potential novel drug targets for major diseases.
Introduction
The vast majority of our current understanding of the genetic architecture of the human proteome is derived from populations of European ancestry. This "ancestry gap" creates a significant blind spot in biomedical research, potentially overlooking biological pathways and drug targets that are more prevalent or unique to other global populations. Proteins are the primary targets for most therapeutic interventions; therefore, mapping the genetic variants that influence protein levels—known as protein quantitative trait loci (pQTLs)—is a critical step in drug discovery. By leveraging Mendelian Randomization (MR), researchers can distinguish causal protein-disease relationships from mere correlations. This study addresses the need for ancestral diversity by presenting a large-scale proteogenomic map of a Chinese population.
Methods
The study utilized the China Kadoorie Biobank (CKB), measuring 1,451 proteins in the plasma of 3,974 participants using the Olink Explore platform. The researchers performed a genome-wide association study (GWAS) to identify pQTLs and employed SuSiE (Sum of Single Effects) fine-mapping to pinpoint causal genetic variants within credible sets.
To assess the impact of ancestry, results were compared against the UK Biobank Pharma Proteomics Project (UKB-PPP), which includes approximately 50,000 individuals of primarily European descent. Causal inference was performed using MR and colocalisation analysis (via the coloc package) to ensure that the genetic signals driving protein levels and disease risk were identical, thereby minimizing false positives. Phenotype data were sourced from the GWAS Catalog, focusing on East Asian (EAS) and multi-ancestry datasets.
Key Insights
Ancestry-Specific Genetic Architecture
A significant finding of this study is that approximately 31% of the fine-mapped genetic signals in the Chinese cohort did not overlap with those identified in European populations. This suggests that the "genetic control panel" for protein expression varies substantially across ancestries. Despite a sample size ten times smaller than major European studies (4,000 vs. 50,000), the researchers identified pQTLs for 74% of the proteins and discovered unique signals (e.g., HARS1 and CDH1) that were entirely absent in the larger European datasets.
Pleiotropy and the ABO Locus
The study highlighted the ABO locus, which determines blood type, as a "master regulator." This locus was found to be highly pleiotropic, influencing the levels of 133 different proteins, underscoring its broad biological impact beyond simple blood group determination.
Results and Therapeutic Implications
The research identified 22 protein-disease pairs with strong evidence of causality, 20 of which had not been reported in previous European-centric studies.
- Novel Drug Targets: Nine potential novel targets were identified. A standout discovery is GP2 (Glycoprotein 2) as a target for Type 2 Diabetes. The genetic association between GP2 and diabetes is robust in East Asians but notably absent in European populations.
- Drug Repurposing: The analysis revealed that HSPA1A (a heat shock protein) exerts a protective effect against Takayasu arteritis, a rare inflammatory vascular disease. Given that drugs targeting HSPA1A are already in development for other indications, this represents a significant opportunity for drug repurposing.
- Validation of Known Targets: The study successfully validated the link between MSMB and prostate cancer, a relationship for which the drug Tigapotide is already in clinical trials.
Conclusion
This study demonstrates that ancestral diversity is not merely a matter of scientific inclusivity but a functional necessity for comprehensive drug discovery. By expanding proteogenomic research to East Asian populations, the authors identified novel therapeutic targets and repurposing opportunities that were hidden in European-only studies. While larger East Asian cohorts are needed to identify trans-pQTLs with smaller effect sizes, these findings underscore the urgent need for global representation in genomic medicine to ensure that the benefits of precision medicine are accessible to all populations.
Summary updated