Every gene needs a signal that says “start here.” Now scientists have decoded one of the most important of these — a DNA element called the initiator — using artificial intelligence.
Researchers at the University of California, San Diego, led by James T. Kadonaga with graduate researcher Torrey Rhyne-Carrigg, set out to identify the initiator’s DNA signature: the pattern that marks where a gene’s information begins to be read and turned into functional products. Using high-throughput sequencing to measure gene activity across roughly 500,000 DNA sequence variants, then training a machine-learning model on the results, they pinned down the characteristic pattern — and found the initiator in about 60% of human genes. The work was published in Genes & Development (2026).
Why the ‘on switch’ matters
Genes must switch on in the right place, at the right time, for the body to develop and stay healthy. When that control goes wrong, disease — including cancer — can follow. “These AI models were found to provide, for the first time, strong predictions of the presence or absence of the initiator in human genes,” Kadonaga said.
Why it matters
Being able to predict where this on-switch sits — and how mutations might disrupt it — gives researchers a sharper tool for understanding gene regulation and anticipating how genetic changes contribute to disease. It is fundamental science, but the kind that underpins future diagnostics and therapies.