“I am curing diseases with AI.”
For Dr. Malick Njie, that statement is neither marketing language nor hype—it is the foundation of a mission placing artificial intelligence at the center of medicine’s future.
Njie, a Gambian neuroscientist and biotech entrepreneur, founded Ecotone AI, building foundational AI models capable of reading the human genome “as a first language.” His work sits at the intersection of genetics, neuroscience, and machine learning—one of healthcare’s most transformative frontiers.
Njie’s trajectory reflects both scientific depth and entrepreneurial ambition. He trained in a Nobel Prize–winning lab at Columbia University, earned a PhD in Neurobiology and Neuroscience from the University of Florida, and is a two-time National Science Foundation award recipient. Before Ecotone, he founded Genetic Intelligence, later renamed Genetic Leap, which became a reported $409 million partner of Eli Lilly and Company.
Through Ecotone, Njie is attempting something even more ambitious: building AI systems that can pinpoint precise genetic mutations behind rare inherited diseases and guide scientists toward cures. Rare genetic diseases now number around 10,000, collectively affecting roughly one in eight people worldwide—from childhood blindness to neurological and developmental disorders. The challenge is scale: the genome contains about three billion genetic “letters,” and finding a single faulty mutation has historically taken years of lab work and massive datasets.
“We think this is a world that could only occur with the use of artificial intelligence to help us understand the human genome,” Njie said.
At the center of Ecotone’s work is dnaSORA, described as the world’s first DNA-reading diffusion transformer. Inspired by the same AI architecture powering image and video tools like Midjourney and OpenAI’s Sora, dnaSORA differs from sequential “autoregressive” systems such as ChatGPT. Diffusion transformers learn by gradually degrading information and reconstructing it—like blurring an image into noise and rebuilding it. Njie argues this approach suits genomics, since genomes are multidimensional and nonlinear.
“It’s taking the entire information space at once,” he explained. “We choose diffusion architectures because of their capacity to see long-distance relationships on both sides.”
In practice, dnaSORA reorganizes DNA into structured representations that allow researchers to identify disease-causing mutations with greater precision than conventional approaches. This helps overcome a major bottleneck: the shortage of usable genomic data. In the U.S., a disease is classified as rare if it affects fewer than 200,000 people, and many conditions touch only a few thousand globally.
AI models typically need tens of thousands of examples to learn effectively, Njie noted, yet many rare disease databases hold only dozens or hundreds of genomes. To close that gap, Ecotone uses diffusion transformers to generate synthetic genomic datasets—artificial genomes that closely resemble real ones while remaining unique. The company is working with St. Jude Children’s Research Hospital, which holds one of the world’s largest sickle cell anemia repositories: 807 fully sequenced genomes. “We need tens of thousands more,” Njie said, describing plans to use those as seed data to generate thousands more synthetic genomes. Ecotone believes this could provide “nearly unlimited synthetic training data at negligible compute costs,” while easing genomic-privacy concerns.
The implications extend beyond diagnosis. Combined with gene-editing tools like CRISPR, which received its first regulatory approvals in 2023, AI systems such as dnaSORA could eventually support targeted interventions to correct faulty genes. Njie emphasizes that Ecotone does not intend to become a pharmaceutical company, but rather a foundational AI layer supporting CRISPR developers and biotech firms.
This reflects a broader shift in healthcare: reports indicate that Anthropic, the company behind Claude, acquired biotech startup Coefficient Bio in a roughly $400 million deal and has expanded into healthcare through Claude for Healthcare and Claude for Life Sciences. OpenAI and others are making similar moves.
Yet in this global race, Africa remains both central and underrepresented. The continent holds the most genetically diverse populations in the world, but contributes less than 3% of global genomic sequencing data. This imbalance has limited precision medicine, since many treatments are built on datasets drawn from Europe and North America.
“For too long, the absence of locally derived genomic data has limited how we prevent, diagnose, and treat disease,” said Dr. Abasi Ene-Obong, founder of Syndicate Bio. Patients often receive less-tailored treatments, hereditary diseases go undiagnosed, and cancer care relies on generalized protocols. Nigeria alone recorded more than 260,000 new cancer cases in 2022, with only 27 treatment centers.
Sequencing the first human genome took 13 years and cost roughly $2.7 billion. By 2024, that cost had fallen to around $200 per genome, yet local genomic infrastructure remains uneven across Africa.
Founded in Nigeria in 2023, Syndicate Bio is trying to close that gap by combining sequencing, AI, and clinical care through its Direct by Syndicate Bio platform. Partner labs in over 20 Nigerian cities feed samples to analysis facilities in Lagos. Each genome can generate up to 100 gigabytes of data, processed by AI into reports covering hereditary cancer testing, tumor profiling, pharmacogenomics, and prenatal screening.
“In Nigeria, we’ve not really been treating cancers from an actionable perspective because we’ve not had the tools,” Ene-Obong said. “But with this now, we’ll be able to know exactly what is driving that cancer.”
That urgency is visible on the clinical front lines. Dr. Abubacarr Jah, Associate Professor of Urology and Medical Director at Sharab Hospital, who also lectures at the University of The Gambia and teaches at Edward Francis Small Teaching Hospital, treats cancer patients and trains medical students. He says the core challenge across Africa is late detection.
“Not all cancers manifest seriously, and that is why people tend to ignore the symptoms,” he explained, noting that prostate and cervical cancers can develop even in people who appear healthy. Screening rates in Gambia remain very low due to cost, unlike wealthier countries where it is routine. Many patients turn first to traditional treatments, delaying hospital visits.
“We tend to see cancers late, and in advanced countries, they see cancers early,” Jah said. “Early solid cancers, such as prostate or breast, are curable in Gambia through surgery, as we have the facilities and expertise.”
The real gap emerges with advanced cancers requiring chemotherapy or radiotherapy, neither widely available in Gambia. Radiotherapy is now accessible in Dakar, Senegal. Chemotherapy and cancer drugs are not classified as essential medicines by the Gambian government, so public hospitals depend on donations. Some private clinics stock the drugs, but affordability remains a major barrier.
Jah recalled ordering cancer drugs that patients could not afford, forcing him to donate the medication to Edward Francis Small Teaching Hospital in Banjul before it expired. “These medicines are so expensive because they cost thousands of pounds,” he said, adding that patients who die because they cannot afford radiotherapy in Dakar represent “the tip of the iceberg.” For every patient who makes the trip, three or four others cannot.
“Our government cannot afford to buy these cancer drugs and equipment, as they are not part of the essential medicines in this country,” Jah explained.
He is hopeful that Gambia’s National Health Insurance Scheme could eventually make treatment affordable, noting that even in wealthy countries, cancer care is largely unaffordable without insurance. Prostate cancer treatment costs around D15,000 in Gambia, versus roughly £6,000 in the UK—cheaper in absolute terms, but still out of reach for most Gambians.
On AI’s potential, Jah believes Africa needs the technology more urgently than the West and is preparing his students accordingly through his work with the West African College of Surgeons.
“Let us not be left behind in terms of AI, as these are cheaper technologies that we can use to tackle diseases and improve our healthcare system in Africa,” he said. “AI will provide us with cheaper, faster, and more effective means to find cures for our people. And who needs it more than us, the poor people?”
Gambia has no national policy on AI precision medicine, though individual doctors are exploring it.
Significant obstacles remain concerns over data ownership, privacy, and equitable access are growing alongside AI-driven genomics, and the World Economic Forum has urged ethical safeguards to prevent exclusion. Many African health systems still lack sequencing capacity and trained specialists, and Gambia has no chemotherapy facility.
Still, momentum is unmistakable. Africa’s genetic diversity is increasingly seen as a scientific asset. Syndicate Bio plans to expand into digital pathology and newborn screening; Ecotone aims to build models capable of understanding unexplored regions of the genome.
“In a year or two, I hope we start scratching at those words that we don’t know,” Njie said, referring to the vast stretches of genetic code still unmapped. Outside the lab, he describes himself as a novice kitesurfer, gardener, and audiophile—a reminder that behind the algorithms are people driven by curiosity.
The story of genomics in Africa is no longer defined only by missing data and infrastructure gaps. It is increasingly shaped by innovation and determination to ensure the continent plays a central role in the future of medicine—one that could bring earlier diagnoses, targeted treatments, and health systems that are predictive, personalized, and data-driven.
