Deepmind's AlphaGenome Atlas predicts effects of 9B DNA changes
deepmind's alphagenome atlas predicts effects of 9 billion possible single-letter dna changes in human genome to help researchers identify disease-linked variants.
google deepmind's alphagenome atlas predicts effects of 9 billion possible single-letter changes in human genome. it builds on ai model alphagenome, introduced in 2025, which reads dna stretches 1 million letters long and predicts gene expression, regulatory protein binding, and gene transcript splicing. the dataset includes 27,000 prediction values per variant and a single number, alphagenome variant impact score (avi), to simplify analysis. avi combines alphagenome predictions with protein model alphamissense and evolutionary conservation measures.