By combining studies in flies with artificial intelligence (AI), researchers in the US have identified alterations in a gene that may cause a rare neurodevelopmental condition in ten patients. 

Researchers from Baylor College of Medicine, Texas, used AI-MARRVEL, an AI model combining clinical information and genetic analysis to identify alterations potentially involved in disease, to analyse the results of the genetic sequencing of a child with an undiagnosed developmental condition. The AI tool found a rare alteration in a gene called BRSK1 that had been missed by standard analysis.  

After sharing the results, the researchers found additional people with alterations in the same gene. They studied ten affected individuals from seven unrelated families. All showed some degree of developmental delay, with features including delayed speech and language, intellectual disability, autism spectrum disorder, anxiety, attention deficit and low muscle tone, and two also had seizures. 

By genetically deleting the equivalent gene in fruit flies, the team discovered that BRSK1 is involved in the communication between neurones. Flies missing the gene had difficulty moving, were more sensitive to stressors that cause seizures and lived shorter lives. These features were largely reversed when the normal human gene was introduced, indicating that BRSK1 and its fly counterpart have similar roles. However, when the researchers introduced versions of the human BRSK1 gene carrying three alterations found in the patients, the flies only partially recovered.  

“This work improves our understanding of the genetic causes of neurodevelopmental disorders and highlights the power of combining AI-driven gene discovery with experimental studies in model organisms to uncover new rare diseases and their underlying biology,” said Hugo Bellen, senior author of the study published in the American Journal of Human Genetics and a researcher at Baylor College of Medicine.