When Matthew Might's infant son Bertrand began showing severe developmental delays, frequent seizures, and an inability to produce tears at just six months old, the computer science professor embarked on a four-year diagnostic odyssey that would transform his career and reshape how scientists approach rare genetic diseases. Today, Dr. Might leads the Hugh Kaul Precision Medicine Institute at the University of Alabama-Birmingham, where he uses artificial intelligence to help families facing similar medical mysteries find answers and potential treatments .

What Makes NGLY1 Deficiency So Difficult to Diagnose?

Bertrand's case was groundbreaking in the worst possible way. When exome sequencing finally revealed the diagnosis after years of searching, he became the world's first confirmed patient with NGLY1 deficiency, a novel autosomal recessive disorder affecting the endoplasmic reticulum-associated degradation pathway. The condition causes neurological dysfunction, abnormal tear production, and liver disease. The inability to produce tears, a condition called alacrima, became the telltale sign that would help other families recognize the disorder .

"That was the beginning of a diagnostic odyssey that lasted 4 years, and it took the invention of exome sequencing to finally resolve it," explained Dr. Might, director of the Hugh Kaul Precision Medicine Institute at the University of Alabama-Birmingham.

Dr. Matthew Might, Director of the Hugh Kaul Precision Medicine Institute at the University of Alabama-Birmingham

What started as a personal tragedy became a global mission. Using social media and a strategically written blog post designed to highlight the key symptoms, Dr. Might found other patients within just two weeks. Today, over 130 confirmed cases of NGLY1 deficiency have been identified worldwide, all connected through an online family group that Dr. Might helped establish .

How Is AI Transforming the Search for Rare Disease Treatments?

Dr. Might's pivot from computer science to precision medicine gave him a unique advantage. He began building AI platforms capable of sifting through scientific literature, integrating new research findings, and identifying potential therapies that might work for genetic disorders. This approach proved remarkably effective for his own son's condition .

When Harvard researcher Gary Ruvkun published a paper showing that NGLY1 was essential for deglycosylating functional proteins like NRF1, it opened new therapeutic possibilities. Dr. Might's AI platform analyzed this discovery and recommended sulforaphane, a sulfur-rich compound known to activate the NRF2 pathway and promote cellular detoxification. The recommendation made molecular sense for Bertrand's specific genetic disorder .

"What's really changing it right now is generative AI. That's relevant because it works on the precision diagnostic side when we're trying to figure out whether particular mutations are actually driving a disease. On the therapeutic side, it helps us figure out if an existing drug can be repurposed for a particular disorder," stated Dr. Might.

Dr. Matthew Might, Director of the Hugh Kaul Precision Medicine Institute

The results were tangible. At one point, Bertrand was suffering more than 100 epileptic seizures per day. Through AI-guided treatment exploration and medical intervention, Dr. Might gradually gained control over those episodes, and Bertrand often experienced days without a single seizure, even shortly before he passed away in 2020 at age 12 .

Ways AI Is Accelerating Rare Disease Research and Treatment Discovery

• Drug Repurposing: AI platforms can rapidly analyze existing approved medications to identify which ones might work for rare genetic disorders, avoiding the lengthy and expensive process of developing entirely new drugs from scratch.
• Literature Integration: Generative AI systems can sift through thousands of scientific papers, extract relevant findings, and identify therapeutic connections that human researchers might miss due to the sheer volume of published research.
• Precision Diagnostics: AI helps determine whether specific genetic mutations are actually driving a disease, enabling more accurate diagnoses and targeted treatment approaches tailored to individual patients.
• Community Building: AI-powered tools and social media strategies help researchers identify and connect patients with rare conditions, creating global networks that accelerate research and provide emotional support.

Dr. Might emphasizes that AI should never replace human medical judgment. He cautioned that when using AI systems for medical advice, patients and families should always consult with qualified clinicians to ground AI recommendations in clinical reality and individual patient circumstances .

What Role Is the Grace Science Foundation Playing in NGLY1 Research?

The foundation, started by Matt and Kristen Wilsey after their daughter Grace was diagnosed with NGLY1 deficiency, has become a major force in advancing research for this ultrarare condition. Based in Menlo Park, California, the foundation focuses on animal models, cell and systems biology, chemistry, genomics, stem cells, and therapy design and development. Since its founding, it has funded 20 research teams across the United States, Canada, Germany, Italy, and Japan .

Dr. Might's work with the foundation reflects a broader shift in how rare disease research operates. Rather than waiting for pharmaceutical companies to develop new drugs, researchers are now using AI to creatively reuse existing approved medications in ways that make molecular sense for specific genetic disorders. This approach is faster, less expensive, and often more practical for the vast majority of rare genetic diseases .

"We're looking for ways to creatively reuse the existing arsenal of approved drugs in ways that make molecular sense for a given genetic disorder. We have whole new techniques that we didn't have even 5 years ago," noted Dr. Might.

Dr. Matthew Might, Director of the Hugh Kaul Precision Medicine Institute

Despite recent advances in AI and precision medicine, children with NGLY1 deficiency generally die in their mid to late teens, though some have lived significantly longer. While it remains unclear whether AI will eventually lead to a cure for this ultrarare condition, Dr. Might's work demonstrates how artificial intelligence can accelerate the discovery process, improve quality of life for patients, and build hope for families facing genetic mysteries. His journey from computer scientist to precision medicine pioneer shows that sometimes the most transformative innovations come from personal tragedy combined with technical expertise and unwavering determination .