Researchers have developed a cutting-edge artificial intelligence (AI) approach that significantly speeds up the identification of genes linked to neurodevelopmental disorders such as autism spectrum disorder, epilepsy, and developmental delay. This powerful computational tool has the potential to fully map the genetic landscape of these conditions, which is essential for making accurate molecular diagnoses, understanding disease mechanisms, and creating targeted therapies. The study, which introduces this novel AI methodology, was published in the American Journal of Human Genetics.

"Although researchers have made major strides identifying different genes associated with neurodevelopmental disorders, many patients with these conditions still do not receive a genetic diagnosis, indicating that there are many more genes waiting to be discovered," said first and co-corresponding author Dr. Ryan S. Dhindsa, assistant professor of pathology and immunology at Baylor College of Medicine and principal investigator at the Jan and Dan Duncan Neurological Research Institute at Texas Children's Hospital.

Traditionally, the discovery of disease-associated genes involves sequencing the genomes of individuals affected by these disorders and comparing them to the genomes of unaffected individuals. While this method has yielded important insights, gaps remain in identifying all relevant genetic contributors to neurodevelopmental conditions.

"We took a complementary approach. We used AI to find patterns among genes already linked to neurodevelopmental diseases and predict additional genes that might also be involved in these disorders," Dhindsa explained.

The research team applied AI to analyze gene expression data obtained from the developing human brain at the single-cell level. By doing so, they were able to identify patterns in gene activity that could indicate an association with neurodevelopmental disorders. "We found that AI models trained solely on these expression data can robustly predict genes implicated in autism spectrum disorder, developmental delay, and epilepsy. But we wanted to take this work a step further," Dhindsa said.

To improve their predictive models, the researchers incorporated more than 300 additional biological features. These included measures of how intolerant certain genes are to mutations, their interactions with other known disease-related genes, and their functional roles within various biological pathways. By integrating this extensive dataset, they created AI models with exceptionally high predictive value.

"These models have exceptionally high predictive value," Dhindsa said. "Top-ranked genes were up to two-fold or six-fold, depending on the mode of inheritance, more enriched for high-confidence neurodevelopmental disorder risk genes compared to genic intolerance metrics alone. Additionally, some top-ranking genes were 45 to 500 times more likely to be supported by the literature than lower ranking genes."

These findings suggest that AI-driven models can serve as powerful analytical tools for validating genes that have begun to emerge from sequencing studies but currently lack sufficient statistical evidence to be definitively linked to neurodevelopmental conditions.

"We see these models as analytical tools that can validate genes that are beginning to emerge from sequencing studies but don't yet have enough statistical proof of being involved in neurodevelopmental conditions," Dhindsa said. "We hope that our models will accelerate gene discovery and patient diagnoses, and future studies will assess this possibility."

The researchers believe that this AI-based approach will not only help identify new risk genes but also refine the classification of genes that contribute to neurodevelopmental disorders. By enabling a more precise understanding of genetic influences on these conditions, the AI models may pave the way for more personalized diagnostic and therapeutic strategies in the future.

This study represents a collaboration among multiple institutions and researchers. Blake A. Weido, Justin S. Dhindsa, Arya J. Shetty, Chloe F. Sands, Slavé Petrovski, Dimitrios Vitsios, and co-corresponding author Anthony W. Zoghbi contributed to the research. The authors are affiliated with Baylor College of Medicine, the Jan and Dan Duncan Neurological Research Institute at Texas Children's Hospital, AstraZeneca, and the University of Melbourne.

The development of this AI-powered gene prediction tool highlights the increasing role of machine learning in medical research. AI is proving to be an invaluable asset in uncovering complex biological patterns that may otherwise remain undetected through traditional research methods. As technology continues to advance, AI-driven approaches may significantly enhance the ability to diagnose and treat a wide range of genetic disorders, ultimately improving patient outcomes.