9 Billion Genomic Variants Now Mappable with Google DeepMind's AlphaGenome Atlas

Google DeepMind has released AlphaGenome Atlas, a massive dataset of precomputed predictions for every possible single-nucleotide variant in the human genome. This includes around 9 billion single-letter changes. The release also introduces the AlphaGenome Variant Impact (AVI) score, which ranks variants by predicted impact.

The AVI score combines AlphaGenome's regulatory predictions with AlphaMissense, DeepMind's model for protein-altering variants. This allows researchers to evaluate both coding and non-coding regions of the genome. The dataset also includes per-variant feature attributions and a genome-wide motif collection.

AlphaGenome Atlas is available as a free web portal for academic use and through the AlphaGenome API. Commercial access on Google Cloud is listed as 'coming soon'.