A study published in Nature Genetics has used AI-based computational models to map how genes interact in the brain to contribute to schizophrenia risk, identifying 766 associated genes, 641 of which had not appeared in previous transcriptomic analyses. Rather than a single faulty gene, schizophrenia appears to stem from hundreds of genetic variants working together in coordinated networks, and this study offers one of the most detailed pictures yet of that architecture, potentially aiding future research into treatments.

The research analysed genetic data from more than 102,000 people alongside brain tissue samples from six brain regions, drawing on work from the Lieber Institute for Brain Development, the University of Bari and numerous psychiatric centres worldwide. Many of the newly identified genes were found through long-range genetic regulatory signals, reinforcing the idea of an interconnected genetic network rather than isolated causes. The World Health Organization estimates schizophrenia affects around 23 million people globally, roughly one in every 345, with symptoms including hallucinations, delusions, social isolation and cognitive difficulties reflecting the disorder's underlying biological complexity.