We would like to congratulate Chen-Yang Su, PhD student of the McGill CERC in Genomic Medicine, who was awarded the Best Oral Presentation at the Cardiometabolic Health, Diabetes and Obesity Research Network (CMDO) Annual Symposium on Health Research and Artificial Intelligence on October 24th, 2024.
You will find the abstract of his presentation below:
Multi-ancestry proteome-phenome-wide Mendelian randomization offers a comprehensive protein-disease atlas and potential therapeutic targets
Authors
Su Chen-Yang, MSc *[1,2], van der Graaf Adriaan, PhD [3], Zhang Wenmin, PhD [4], Selber-Hnatiw, Susannah, MSc [2,5], Richards Brent, MD, MSc [5,6,7,8,9], Flannick Jason, PhD [10,11], Zhou Sirui, PhD [2,5], Mooser Vincent, MD [2,5], Lu Tianyuan, PhD [12,13,14,15], Yoshiji Satoshi, MD, PhD [2,5,6,10,11]
Objective: Circulating proteins influence disease risk and are valuable drug targets. To increase the discovery of protein-phenotype associations and identify therapeutic targets for diverse populations, we conducted multi-ancestry proteome-phenome-wide Mendelian randomization (MR), followed by sensitivity analyses and druggability assessment.
Methods: We performed two-sample MR, analyzing 2,265 unique proteins from two platforms (SomaScan v4 and Olink Explore 3072)—2,110 proteins from four European cohorts (n = up to 35,559 individuals), 1,144 from two African cohorts (n = up to 1,871), and 581 from a novel East Asian cohort (n = 1,823). We curated the largest GWAS for 179 traits in Europeans, 26 in Africans, and 206 in East Asians. To reduce horizontal pleiotropy, we ensured instruments were cis-pQTL for the protein of interest and had the highest Open Targets variant-to-gene score. We performed sensitivity analyses including colocalization with PwCoCo and SharePro, a novel method we recently developed. We evaluated shared causal effects of prioritized proteins across ancestries and assessed overlap with the druggable genome, Open Targets, and DrugBank.
Results: We tested 726,035 protein-phenotype pairs in Europeans, 33,078 in Africans, and 115,352 in East Asians. Notably, 119 proteins were instrumentable only in Africans and 17 proteins only in East Asians due to allele frequency differences that are common in these ancestries but rare in Europeans. We identified 3,949, 56, and 325 unique protein-phenotype pairs in European, African, and East Asian populations, respectively. To showcase our findings, we illustrated the causal role of IL1RL1 in inflammatory bowel diseases, which was supported by multiple lines of evidence.
Conclusion: This study provides the largest multi-ancestry atlas of protein-disease associations, expanding insights into disease etiology and opportunities for prioritizing therapeutic targets. Results are available at the Common Metabolic Diseases Knowledge Portal (https://hugeamp.org/r/protein_mr_atlas).