The University of Cape Town has won $1.6 million (about R26.4 million) from Google.org to build an AI platform mapping the genetic variants that determine how African patients metabolise medicines.
UCT is one of 17 organisations selected worldwide for the Google.org Impact Challenge: AI for Science, from more than 3,000 applicants across 30 countries. The 17 share $36 million.
The gap it addresses
“Medicines do not work the same way in everyone, yet the evidence that guides prescribing in Africa has largely been generated elsewhere,” said Professor Kelly Chibale, who holds the Neville Isdell Chair in African-centric Drug Discovery & Development and founded UCT’s H3D centre.
“Africa is home to the greatest human genetic diversity on Earth, and that diversity must be at the centre of the science that shapes how medicines are used.”
Most pharmacogenomic guidelines have been built almost entirely from European and East Asian data, leaving African clinicians prescribing critical medicines without population-specific evidence.
What AfriPGx-AI will do
The platform will catalogue genetic variation across more than 50 drug-metabolising genes in African populations and predict which variants are most likely to alter how medicines are processed.
The strongest predictions then go to the laboratory for testing — what Chibale describes as “a tight loop between computational prediction and experimental validation that focuses effort where it matters most.”
It starts with tuberculosis, HIV and malaria — the diseases carrying the heaviest burden on the continent, and where imported dosing guidance is least likely to apply. AI predictions are combined with experimental data and mathematical modelling to build evidence for safer medicine selection and dosing.
Two decisions that determine reach
Everything is being released openly. Software, AI models and curated datasets will be published, with findings shared through international research databases.
The decision-support tools are being built for low-bandwidth settings, so they can be used where they are most needed rather than only in well-connected teaching hospitals.
Those two choices decide whether the work reaches a district clinic or stays in the literature.
Because the methods are not tied to one population or one class of medicine, they can extend to other underrepresented populations and to cancer and cardiovascular treatments.
The team, and what it builds
The project involves Chibale alongside Dr Musalula Sinkala of the Department of Integrative Biomedical Sciences’ Division of Computational Biology; Dr Mwila Mulubwa and Dr Mathew Njoroge, senior and chief investigators at H3D; and Professor Jonathan Shock, interim director of the UCT AI Initiative.
The award will also train African postdoctoral fellows and MSc and PhD students in AI, bioinformatics and pharmacogenomics.
“This award allows us to bring AI and laboratory science together to close that gap, and to do so in Africa, led by African scientists, with tools that are open to all,” Chibale said.
Leslie Yeh, director of scientific progress at Google.org, said UCT was “bringing together AI and deep scientific expertise in ambitious, creative ways.”
No timeline, researcher numbers or target date for clinical availability were disclosed.
Based on reporting by Thami Nkwanyane for UCT News.
