# Google's AlphaGenome 2 maps disease at the cellular level

> DeepMind's new AI model predicts how DNA mutations affect individual cell types, potentially revealing disease causes and treatment paths.
- Title: Google's AlphaGenome 2 maps disease at the cellular level
- Summary: DeepMind's new AI model predicts how DNA mutations affect individual cell types, potentially revealing disease causes and treatment paths.
- Keywords: google deepmind, alphagenome 2, ai model, genetics, cell biology, drug development, disease research, The Verge
- Source: The Verge — https://theverge.com/ai-artificial-intelligence/1031524/google-deepmind-alphagenome-2-genetic-ai-model
- Author: Robert Hart
- Published: 2026-11-11T15:00:00+00:00
- Read time: 1 min
- Topics: google deepmind, alphagenome 2, ai model, genetics, cell biology, drug development, disease research, technology, culture
## DeepMind's new genetics model
Google DeepMind has unveiled AlphaGenome 2, an AI model designed to show how changes in DNA affect different cells.The model offers a much finer-grained view than its predecessor, potentially revealing disease-relevant effects in specific cells.
## Why cell type matters
Nearly every cell in the human body carries the same DNA, but different cell types switch different genes on and off.
## Training on millions of cells
AlphaGenome 2 was trained on more than one hundred million individual cell profiles across sixty eight tissues.
## Predicting without retraining
AlphaGenome 2 can make predictions about new types of cells and biological states without needing to be retrained.
## Connecting genetics to disease
Adam Gayoso, a research scientist at Google DeepMind, says the breakthrough is tying human genetics to cell biology.

> tying human genetics to cell biology
## Why specificity matters for drugs
Žiga Avsec, Google DeepMind's genomics lead, told The Verge that specificity is vital for drug development. When you develop drugs, you are targeting specific proteins and specific cell types, making this kind of granularity essential.

> When you develop drugs, you're targeting specific proteins and specific cell types
## Computational trade-offs
AlphaGenome 2 is not replacing its predecessor yet. Its more detailed predictions come at a computational cost and are more expensive to produce than those of AlphaGenome, making it impractical for genome-wide atlases.
## Available now for research
Google is making AlphaGenome 2 available for non-commercial research via the AlphaGenome API from today, with a limited private preview for commercial research.
## Key takeaway

Google DeepMind released AlphaGenome 2, an AI model that predicts how genetic variants affect specific cell types, offering finer detail than previous models to help identify disease causes and guide drug development.