AI Maps 766 Genes Tied to Schizophrenia

Schizophrenia affects about 23 million people worldwide and carries a substantial genetic component, but conventional studies have largely tracked nearby DNA variants, leaving long-range regulation and gene-network effects poorly understood. An international team involving the University of Bari Aldo Moro and the Lieber Institute for Brain Development used advanced computational models to connect genetic variation with activity across brain networks, an approach that could sharpen understanding of disease biology and help researchers prioritize drug targets for precision treatment.

The study, published in Nature Genetics on June 22, 2026, analyzed genetic data from 102,613 people across 62 Psychiatric Genomics Consortium cohorts and RNA-sequencing data from six post-mortem brain regions. The researchers combined two models, INGENE and MODULE, with conventional predictors to capture distant regulatory effects. They identified 766 genes significantly associated with schizophrenia, including 641 not detected in previous transcriptome-wide analyses, supporting a model in which coordinated gene networks, rather than any single gene, shape risk.