Google DeepMind unveils AlphaGenome 2 to model cellular impacts of DNA mutations

Researchers at Google DeepMind have announced AlphaGenome 2, the successor to their previous genome-predicting AI model. Trained on more than 100 million individual cell profiles across 68 tissues, the new model predicts how genetic variants alter gene expression within specific cell types and states. This granular view allows scientists to connect human genetics directly to cell biology. The development addresses the challenge that the same genetic variant can have entirely different effects depending on the type and state of a cell. According to the company, AlphaGenome 2 identifies nearly twice as many gene-expression-changing variants as its predecessor and can make predictions for new cell types without retraining, potentially aiding future laboratory research and drug discovery.