Loading Curie Briefs...
Getting the latest healthcare news for you
Getting the latest healthcare news for you
AI just gave genomics researchers a massive upgrade. Google DeepMind's AlphaGenome Atlas catalogs AI-generated predictions for over 9 billion possible DNA changes across hundreds of human cell types — all searchable without writing a single line of code. The tool also introduces a new variant impact score that could help scientists zero in on disease-causing genetic variants faster than ever before.
AI just gave genomics researchers a massive upgrade.
Google DeepMind has launched AlphaGenome Atlas, a one-petabyte, freely accessible resource that maps the predicted molecular effects of more than 9 billion possible single-letter changes in the human genome. Developed in collaboration with scientists at the Stowers Institute for Medical Research, the Broad Institute, and the University of Exeter, the Atlas covers hundreds of human cell types and tissues — and requires no coding skills to use.
At the heart of the tool is the new AlphaGenome Variant Impact (AVI) score, which combines predictions from AlphaGenome, AlphaMissense, and evolutionary conservation data into a single ranking system for variant impact across both protein-coding and non-coding regions of the genome.
By the Numbers:
Why it matters: Identifying which genetic variants cause disease has long been a bottleneck in genomics research. AlphaGenome Atlas doesn't replace lab experiments — but it can tell scientists where to look first, potentially accelerating rare disease diagnosis, drug target discovery, and our fundamental understanding of gene regulation.
