Approved Research

Genetic prediction of common diseases in diverse populations

Lay summary

It is becoming increasingly feasible to use a person's DNA to predict their risk of common diseases.  However most research to date has been conducted in European ancestry populations, and studies in other populations, particularly those of African ancestry, have been less encouraging.  Genetic prediction therefore has to potential to exacerbate health disparities while improving risk management.  In order to address this discrepancy, it is necessary to both increase the amount of research data available from diverse populations, and to understand and address the fundamental reasons why results from European ancestry do not translate well to the other populations.

This project will use genetic data in the UK Biobank together with datasets from other populations, particularly in African countries including Uganda, Malawi, Kenya and South Africa.  We will evaluate genetic predictors developed in European ancestries across different African populations to understand the factors driving their accuracy, with a focus on heart and kidney disease and related traits such as body-mass index and cholesterol levels.  We will explore the use of predictors developed from African diaspora, including African-ancestry individuals in UK Biobank, and compare their usefulness between different African populations.  We will develop and apply methods to combine information across multiple ancestries to obtain prediction models with global relevance.  In so doing we will establish prospects and guidelines for achieving equitable applications of genetic risk prediction.