Google’s AlphaGenome 2 zooms in on the cellular roots of disease

Google DeepMind announced AlphaGenome 2 on November 11, 2026, describing it as a next-generation DNA sequence model that bridges single-cell biology and human genetics by predicting how genetic variants affect hundreds of human cell types. The model was trained on more than 100 million individual cell profiles across 68 tissues, including over 107 million individual cell profiles from human and mouse samples, using both bulk and single-cell data. According to Google DeepMind, AlphaGenome 2 identifies nearly twice as many variants that change gene expression in cells compared to the previous state-of-the-art model.

The original AlphaGenome predicted DNA variant effects across broad tissues using a parallel approach that processed thousands of experiments at once. AlphaGenome 2 fundamentally redesigns this architecture as a conditional model that predicts DNA effects for one biological condition — defined as a tissue, cell type, or state — at a time. The target condition is used as input at every stage of the architecture, allowing the model to learn how the same DNA sequence is read differently across cell types. This conditional design also enables cross-modality imputation: given a single measurement such as single-cell gene activity, the model can predict unmeasured properties like chromatin accessibility.

AlphaGenome 2 predicts core experimental readouts including gene expression, histone modifications, chromatin accessibility, and DNA methylation, a widely requested addition. The model can generalize to entirely new single-cell experiments without retraining. Scientists can upload a single-cell RNA-seq snapshot from just a few dozen cells; this gene activity is compressed into a cell state representation and used to condition the model on unseen states. Google DeepMind said AlphaGenome 2 can match or outperform models trained or fine-tuned directly on those datasets.

As an example, given unseen single-cell data from a COVID-19 mRNA vaccine study, AlphaGenome 2 distinguished vaccine-activated immune T cells from resting ones and revealed that a variant linked to severe COVID-19 disrupts gene activity more strongly in the activated state — an effect missed in unactivated cells. The company has precomputed AlphaGenome 2 predictions for trait-associated variants across hundreds of human cell and tissue conditions to help researchers connect genome-wide association study variants to single-cell maps without running large-scale computations themselves.

AlphaGenome 2 is available today for non-commercial research via the AlphaGenome API and in private preview for commercial research. Interactive Colab tutorials for expression conditioning are also available. The company emphasized that AlphaGenome and AlphaGenome 2 are intended for research use only and have not been validated or approved for clinical or diagnostic purposes. Community feedback is being sought via a form due by November 30 and a community forum to shape the future roadmap.