Wits University scientists are at the forefront of an ambitious project to analyse hundreds of human genomes from African populations, paving the way to a better understanding of the complex interplay between genetic variation and disease.
Africa is the anatomical home of modern humans and boasts the greatest genetic diversity, yet until now less than 2% of the human genomes sequenced came from people of African origin. Global genomic databases and research studies have been dominated by people of European ancestry.
“It’s a missed opportunity because we don’t have enough data from the people who could provide the most information to advance science and our understanding of health processes,” said Michèle Ramsay, director of the Sydney Brenner Institute for Molecular Bioscience at Wits.
“As a result, the genetic patterns guiding modern medicine are very Eurocentric and poorly predict disease in African populations,” said Ramsay, who is the lead author of a study — published this week in Nature — describing the process of collecting African human genomes.
If we don’t feed the AI algorithms with African data, what we get is going to be biased.
— Michèle Ramsay, director of the Sydney Brenner Institute for Molecular Bioscience at Wits
It is crucial the AI models being developed for tasks such as early disease detection include genetic information from African populations, she said.
“If we don’t feed the AI algorithms with African data, what we get is going to be biased,” she said.
The Assessing Genetic Diversity in Africa (AGenDA) project, run under the Human Heredity and Health in Africa (H3Africa) consortium, identified underrepresented groups spanning eight countries for human whole-genome sequencing.
Covering the roughly 2,000 ethnolinguistic groups in Africa would have required sequencing hundreds of thousands of genomes, so the scientists focused on five population groups — people with Khoe-San hunter-gatherer ancestry; Niger-Congo-speaking people from central and central-southern Africa; Nilo-Saharan speakers; North African and Afro-Asiatic speakers; and people from the African islands in the southwest Indian Ocean.
Samples were collected from people in Angola, the Democratic Republic of Congo, Kenya, Libya, Mauritius, Rwanda, Tunisia and Zimbabwe, but not South Africa, which has previously contributed material for research on the human genome.
AGenDA’s processes for data governance and data sharing are aligned with World Health Organisation guidance, and the data will be submitted to the European Genome-Phenome Archive. The datasets will be openly accessible to the global research community, and a data-access committee will manage requests for access, said Ramsay.
The new data on African genomes is expected to help other scientists deepen their understanding of a wide range of diseases and develop new treatments.
“Because African genomes contain the deepest branches of human genetic history, they help scientists distinguish genetic variants that are ancient from those that arose more recently and help to detect variants that truly influence disease,” said Ramsay.
Would you like to comment on this article?
Sign up (it's quick and free) or sign in now.
Please read our Comment Policy before commenting.