According to CryptoSlate, Google DeepMind unveiled AlphaGenome 2 on November 11, which can predict the effects of DNA alterations by incorporating target cell types or states. DeepMind stated that in the TenK10K evaluation covering 28 primary immune cell types, the model identified 30% of causal variants, compared to 16% for the original AlphaGenome fine-tuned on single-cell data. The company reported these as research benchmark results rather than diagnostic accuracy rates. The report also noted that the new model computationally replicated the previously established ETS2 regulatory mechanism in gut immune cells, rather than discovering it for the first time. The model is available via API for non-commercial research, while commercial research is currently in private preview; it has not been validated or approved for clinical or diagnostic use.