Your DNA contains about 3 billion letters of code, but only 2% of them spell out instructions for making proteins. The other 98% is a vast control panel—switches and dimmers that decide which genes turn on, in which cells, and how loudly. Most inherited diseases, from heart conditions to certain cancers, trace back to glitches in this control panel, not the genes themselves. The problem: finding which glitches matter has been like searching for a faulty wire in a city's electrical grid.
Google DeepMind just released a tool that might change that. AlphaGenome is an AI system trained on human and mouse genetics to predict exactly how mutations scramble gene regulation. It can analyze up to 1 million letters of DNA at once and map out which mutations affect which cell types and biological processes.
"We see AlphaGenome as a tool for understanding what the functional elements in the genome do, which we hope will accelerate our fundamental understanding of the code of life," said Natasha Latysheva, a DeepMind researcher. The distinction matters: most gene-hunting tools focus on the 2% that codes for proteins. AlphaGenome tackles the 98% that controls when and where those proteins get made.
We're a new kind of news feed.
Regular news is designed to drain you. We're a non-profit built to restore you. Every story we publish is scored for impact, progress, and hope.
Start Your News DetoxThe practical payoff could be significant. Researchers can now pinpoint which mutations actually drive disease—and which are just passengers. That clarity opens doors to precision treatments: a drug designed to counteract a specific regulatory mutation, or a gene therapy that switches on the right gene in the right cells. Marc Mansour, a pediatric cancer researcher at UCL, is already using it and describes the shift as a "step change" in finding what drives certain cancers. Gareth Hawkes, a statistical geneticist at the University of Exeter, notes that making sense of that 98% non-coding genome has been a major blind spot until now.
This is early-stage research—the tool is being released to the scientific community for testing and refinement. But the direction is clear: as AI gets better at reading the genome's instruction manual, the path from genetic discovery to treatable disease narrows. The next phase will be watching whether predictions made in the lab translate into treatments that actually work in people.
