Artificial intelligence and advanced DNA sequencing are helping African patients access rare disease diagnoses that were previously out of reach, while simultaneously addressing a critical gap in global genetic databases that have long excluded African genomic data. Geneticist Shahida Moosa at Stellenbosch University in South Africa leads an Undiagnosed Diseases Programme that provides free genomic testing for patients with rare and complex disorders, achieving diagnostic success rates that rival or exceed those in wealthier nations.

Why Are African Patients Underdiagnosed for Rare Diseases?

Families across Africa arrive at clinics carrying years of unanswered medical questions. Children present with seizures, developmental delays, unusual facial features, or extra fingers and toes. Doctors struggle to identify these conditions, leaving families without answers or treatment options. An estimated 100 million Africans are living with rare disorders, including around 4 million people in South Africa alone, yet diagnostic tools and genetic databases have historically been built on non-African populations.

The core problem is straightforward: global genetic databases contain disproportionately little African genomic data. This gap affects both human clinicians and artificial intelligence systems trained largely on non-African populations, making it harder to identify genetic variants that are common in African communities but rare or absent in the databases used to train diagnostic tools.

What Results Is Moosa's Team Achieving?

Moosa's Undiagnosed Diseases Programme is generating striking results. When her team sequenced the exomes, the protein-coding regions of DNA, from 1,000 patients, they achieved a 51% diagnostic success rate, a figure comparable to and in some cases higher than many clinics in wealthier countries. For families, a diagnosis can be transformative. As Moosa explains, many families are relieved to learn that genetics are responsible for their child's condition, rather than something the mother ate or did during pregnancy. A diagnosis also opens the door to treatment, genetic counselling, and better long-term care.

Beyond individual diagnoses, Moosa's work is generating new scientific discoveries. This year alone, her group has published studies demonstrating how RNA sequencing can help identify new variants linked to ciliopathies, disorders affecting tiny hair-like structures on cells, and another identifying a new variant associated with Mabry syndrome.

How Is AI Improving Diagnosis Across Populations?

Before returning to South Africa, Moosa helped develop an AI-based tool that assists clinicians in diagnosing genetic disorders from facial features. Her team now contributes African patient data to the system, helping improve its accuracy across populations that have been historically underrepresented. This feedback loop is critical: AI systems trained on diverse populations perform better for everyone, not just underrepresented groups.

The work extends beyond South Africa. Since 2023, Moosa's efforts have expanded through the Genomics for Health in Africa Cluster of Research Excellence network, a 10-year African-European collaboration involving 18 countries. The initiative is extending genomic testing for undiagnosed rare diseases and familial cancers to countries including Angola, Botswana, and Rwanda.

Steps to Strengthen Genomic Medicine in Underserved Regions

• Expand Genetic Databases: Actively recruit African patients and other underrepresented populations into genomic research to build more inclusive reference databases that reflect human genetic diversity and improve AI model accuracy.
• Integrate Genomics Into Medical Education: Redesign undergraduate and graduate medical curricula to strengthen focus on genomics and genetics, ensuring clinicians in all regions understand how to interpret genetic tests and counsel patients.
• Combine Research, Teaching, and Patient Care: Build programmes that unite clinical diagnosis, graduate student supervision, and research discovery, creating sustainable models that serve patients while advancing science.
• Develop Culturally Appropriate AI Tools: Train diagnostic AI systems on diverse populations and validate their performance across different genetic backgrounds to ensure equitable accuracy.
• Provide Free or Subsidized Testing: Remove financial barriers to genomic testing by offering free services to patients who cannot afford private testing, ensuring rare disease diagnosis is not limited by income.

Alexander Hoischer, a human geneticist at Radboud University Medical Center in the Netherlands, underscores the broader significance of this work. "The inclusion of African diversity is of huge impact to the entire field," he stated. The absence of African genomic data has skewed global understanding of human genetic variation, making it harder to diagnose diseases in African patients and limiting scientific discovery about genetic diversity itself.

"She knows exactly what has to be done, and she is showing that if it is possible to do in South Africa, it is possible elsewhere too," said Hugues Abriel, a molecular geneticist at the University of Bern in Switzerland.

Hugues Abriel, Molecular Geneticist at the University of Bern

Moosa's journey to this work was not straightforward. She began her medical training intending to specialize in pediatrics, but became drawn to the unanswered questions surrounding congenital disorders. After completing clinical training in Johannesburg and earning her PhD from the University of Cologne in Germany, she worked at Harvard and Boston Children's Hospital, where she helped identify eight new genes linked to genetic disorders. Yet that work was not done on African patients. Two years into what was meant to be a five-year stay in Massachusetts, she decided to return to South Africa to focus on the patients and families in her home country.

Since joining Stellenbosch University six years ago, Moosa has built a comprehensive programme combining research, teaching, and patient care. She redesigned undergraduate medical education to strengthen its focus on genomics, supervises graduate students, and runs several clinics for adults, children, and patients with neurogenetic disorders. Her research group uses RNA and short- and long-read genome sequencing to investigate the causes of rare diseases in African patients.

"Dr Moosa is a highly professional and socially engaged clinician, recognised for her kindness and unwavering commitment," noted Ambroise Wonkam, a medical geneticist at Johns Hopkins University School of Medicine in Maryland.

Ambroise Wonkam, Medical Geneticist at Johns Hopkins University School of Medicine

Looking forward, Moosa continues tracking who comes to the clinic and what happens after diagnosis. The evidence she is gathering, she hopes, will help persuade health authorities that rare diseases deserve far greater attention and investment. By demonstrating that high-quality genomic diagnosis is possible in Africa, with success rates matching or exceeding those in wealthier countries, she is making a case that equitable access to genetic medicine is not just ethically necessary but scientifically achievable.