Google DeepMind Launches AlphaGenome 2, Trained on 107M Cell Profiles Across 68 Tissues

Google DeepMind released AlphaGenome 2, a next-generation DNA sequence model trained on over 107 million individual cell profiles across 68 tissues from bulk and single-cell human and mouse samples. The model identifies nearly twice as many variants that alter gene expression as its predecessor, hitting 30% causal variant detection on the TenK10K dataset (vs 16% for AlphaGenome 1) and correct direction 43% of the time (vs 23%). Available today via the AlphaGenome API for non-commercial research, with commercial research in private preview and AlphaGenome 2 Skills coming to Google Antigravity.