Google DeepMind has released AlphaGenome Atlas, a predictive map that catalogues the molecular effects of 9 billion single-letter DNA variants spread across the human genome. The resource is designed to show what happens at the molecular level when one DNA letter is swapped for another, rather than only flagging variants linked to known conditions.
The atlas is built with AI models that infer the likely functional impact of each variant. DeepMind frames it as a step toward making genomic interpretation more comprehensive, since experimental testing of all possible variants would be impractical. The scale — billions of changes — is notable, but the source is a single announcement from DeepMind, so there is no independent assessment of accuracy or clinical utility in the material.
Because only one source is available, the article reflects DeepMind’s own description of the work. The key claim is the breadth of coverage: nearly every possible single-letter change in the human genome has been given a predicted molecular effect. Whether those predictions hold up in practice will require follow-up studies beyond the release.