Development and Validation of Polygenic Risk Scores for Blood Pressure Traits in Continental African Populations
Authors:
- Ebuka OnyenobiEOEbuka Onyenobi
- Michael ZhongMZMichael Zhong
- Opeyemi SoremekunOSOpeyemi Soremekun
- Abram KamizaAKAbram Kamiza
- Romuald BouaRBRomuald Boua
- Tinashe ChikoworeTCTinashe Chikowore
- ACCME Research GroupAGACCME Research Group
- Segun FatumoSFSegun Fatumo
- Ananyo ChoudhuryACAnanyo Choudhury
- Scott HazelhurstSHScott Hazelhurst
- Clement AdebamowoCAClement Adebamowo
- Michèle RamsayMRMichèle Ramsay
Open Access
- 1 June 2025
- journal article
- Published by Wolters Kluwer Health in Circulation: Genomic and Precision Medicine
- Vol. 18 (3) , e005048
Abstract
BACKGROUND: Most polygenic risk scores (PRS) have been developed in European populations, frequently leading to limited transferability across diverse ancestry populations. This study aimed to develop and evaluate PRS for blood pressure (BP) traits in continental African populations and investigate how African genetic diversity influences PRS performance. METHODS: We generated PRS for systolic BP, diastolic BP, pulse pressure, and hypertension. We used a pan-African cohort as the target population and compared single-ancestry and multi-ancestry PRS methods. We compared the performance of African ancestry-derived PRS against multi-ancestry PRS on the entire data set and within South, East, and West African subpopulations. RESULTS: Multi-ancestry PRS demonstrated significantly higher predictive accuracy compared with single-ancestry PRS models. PRS predictive accuracy varied across different African regions, with the highest performance observed in East Africa. In the combined population, the difference in mean BP values between the first multi-ancestry PRS quartile and the top quartile was 6.53 (95% CI, 5.33–7.74), 3.81 (95% CI, 3.09–4.52), and 3.59 (95% CI, 2.84–4.32) mm Hg for systolic BP, diastolic BP, and pulse pressure, respectively. Individuals in the highest PRS risk quartile had odds of hypertension that were 1.47 (95% CI, 1.27–1.69) times greater than those in the lowest risk quartile. CONCLUSIONS: These findings highlight the importance of integrating diverse ancestries in PRS development and accounting for subpopulation genetic variation to improve the predictive accuracy of BP PRS in African populations.Keywords
This publication has 63 references indexed in Scilit:
- Polygenic prediction across populations is influenced by ancestry, genetic architecture, and methodologyCell Genomics, 2023
- Low and differential polygenic score generalizability among African populations due largely to genetic diversityHuman Genetics and Genomics Advances, 2023
- Uganda Genome Resource: A rich research database for genomic studies of communicable and non-communicable diseases in AfricaCell Genomics, 2022
- Genome-wide study on 72,298 individuals in Korean biobank data for 76 traitsCell Genomics, 2022
- Sex Differences in the Prevalence, Outcomes and Management of HypertensionCurrent Hypertension Reports, 2022
- Genome-wide Association Studies in Ancestrally Diverse Populations: Opportunities, Methods, Pitfalls, and RecommendationsCell, 2019
- Regional and Sex Differences in the Prevalence and Awareness of Hypertension: An H3Africa AWI-Gen Study Across 6 Sites in Sub-Saharan AfricaGlobal Heart, 2017
- Polygenic scores via penalized regression on summary statisticsGenetic Epidemiology, 2017
- Modeling Linkage Disequilibrium Increases Accuracy of Polygenic Risk ScoresThe American Journal of Human Genetics, 2015
- A Better Coefficient of Determination for Genetic Profile AnalysisGenetic Epidemiology, 2012