Google’s Atlas of the human genome could pave the way for new treatments
Key Points:
- Google DeepMind has launched AlphaGenome Atlas, an AI tool providing a predictive map of every possible single-letter DNA change in the human genome, aiming to advance biology research and disease treatment development.
- The platform offers predictions on how approximately nine billion potential DNA variants could affect molecular biology, including protein production, making it the most comprehensive catalog of genetic mutation impacts.
- Researchers can access AlphaGenome Atlas via a web portal and related interfaces, and utilize the newly introduced Variant Impact Score (AVI) to prioritize and interpret the effects of genetic variants efficiently.
- The project builds on previous DeepMind models like AlphaGenome and AlphaMissense, extending predictions beyond protein-coding regions to regulatory genome areas that control gene behavior.
- AlphaGenome Atlas is available now for noncommercial research use, with commercial access coming soon on Google Cloud, reflecting Google's broader commitment to applying AI in scientific and medical breakthroughs.