Google Deepmind

AlphaGenome Atlas is a freely accessible database from Google DeepMind that publishes precomputed predictions for the molecular effects of all 9 billion possible single-letter DNA changes in the human genome . It is aimed at geneticists, disease researchers, and computational biologists who want to look up variants without running a model themselves, and it ships with an intuitive website portal, the AlphaGenome API, and as a skill in Google Antigravity . The launch matters because testing that many mutations experimentally is not feasible, and until now the underlying AlphaGenome model had to be run variant by variant to get answers. The core addition on top of last year's AlphaGenome release is the AlphaGenome Variant Impact (AVI) score, which combines the strengths of AlphaGenome and AlphaMissense, DeepMind's model for predicting the impact of protein-altering DNA variants, condensing both models' predictions into a single number so researchers can rapidly rank variants and interpret their molecular effects at the same time . Each variant is also linked to thousands of underlying molecular effect predictions across hundreds of human and mouse cell types and tissues, along with feature attributions that show which biological signals drove the score. The full resource weighs in at a 1-petabyte dataset, more than 30 times larger than the AlphaFold Database . DeepMind is positioning Atlas as the genomics counterpart to what AlphaFold did for protein structure, and reports that external collaborators have already used AlphaGenome Atlas to identify and experimentally verify key variants in unsolved rare disease research and find rare variants associated with common traits . For founders and operators in TechBio, it is worth watching as a new default reference layer that downstream tools, from rare-disease diagnostics to GWAS interpretation pipelines, can build on without hosting the model themselves.