AlphaGenome 2
In brief
AlphaGenome 2 from Google DeepMind predicts how genetic variants disrupt individual cell types, advancing beyond tissue-level averages 1.
Benchmarks and how it works
4 from 1 source
How it worksDiagramWho can use it
AlphaGenome 2 is a new model from Google DeepMind that extends previous tools like AlphaGenome and AlphaGenome Atlas to offer cell type specificity 1.
Does this affect you?
- DevelopersDevelopers can integrate genomic prediction models into pipelines to analyze DNA variants at scale 1.
- Product managersProduct managers can scope healthcare or drug discovery features leveraging advanced genetic prediction models 1.
- FoundersFounders in biotech can leverage cell-type specific predictions to accelerate therapeutic target discovery 1.
- Everyday usersEveryday users do not interact directly with genomic models, though it advances personalized healthcare 1.
How to start
Google DeepMind has not published setup instructions or code for AlphaGenome 2 yet 1.
What you can build
- Cell type variant analysisPredict how genetic variants disrupt individual cell types to pinpoint cellular drivers of disease 1.
Availability
Sources
2 merged into this cardMore from Google

Tool Retrieval in the Gemini API
Gemini's Interactions API now supports on-demand loading of tool parameter schemas. Agents with large tool libraries can keep schemas out of context until needed, cutting input tokens, latency, and selection errors.

Computer use in Gemini app for Mac
Google added computer use to the Gemini app for Mac, letting Gemini automate multi-step tasks across local files, desktop apps, and the web in an isolated background window.
