DeepMind's ultra-powerful AI for genome mapping wows scientists
Google DeepMind says the model was trained on more than 100 million cells and can flag variant effects that bulk methods miss.
- On Wednesday, Google DeepMind announced AlphaGenome 2, an AI model trained on more than 100 million individual cell profiles across 68 tissues to predict how genetic variants affect specific cell types.
- The original AlphaGenome, released last year, relied on bulk samples, leaving researchers unable to pinpoint specific cell types affected by genetic variants. AlphaGenome 2 addresses this limitation by predicting effects within individual cell states.
- Immunologist and geneticist James Lee of the Francis Crick Institute noted the model replicated his 2024 IBD study findings in 30 minutes, identifying the ETS2 gene variant's role in immune cells—a process previously requiring 1.5 years of experiments.
- Starting today, Google is providing access to AlphaGenome 2 for non-commercial research via the AlphaGenome API, with a limited 'private preview' for commercial research applications.
- Avsec stated that future versions could potentially incorporate data 'across multiple species' to broaden biological understanding, while longer-term efforts aim to develop 'virtual cells'—digital simulations designed to mimic biological systems.
13 Articles
13 Articles
Google’s AlphaGenome 2 zooms in on the cellular roots of disease
A graphic representing Google’s AlphaGenome Atlas. | Image: Google Researchers at Google DeepMind have unveiled a new AI model designed to unravel how changes in our DNA affect different cells, giving scientists a more detailed view of a notoriously difficult problem that could hold important clues to the causes of disease - and point the way to new treatments. AlphaGenome 2 is the successor to DeepMind's genome-predicting AI model AlphaGenom…
Google introduces the AlphaGenome 2 DNA sequence model - Root-Nation.com
Google has introduced AlphaGenome 2, a new model designed to predict how DNA sequences and their variants affect different types of human cells. It is intended to become a powerful tool for scientists studying the cellular mechanisms behind disease. The model combines single-cell biology with human genetics, turning static cell atlases into a predictive system […] Post Google Introduces AlphaGenome 2: The Model Predicts How DNA Affects Human Cel…
Google taught AI to read DNA at the level of individual cells – this will help find the roots of diseases
The same DNA for each person does not mean the same effect on different types of cells. The activity of genes in cells is largely individual and subject to disruptions, which can lead to genetic and other diseases. Google’s new […]Read more at Aroged
Google DeepMind launches AlphaGenome 2 to predict DNA effects in single cell types
Genomics Limited, led by Peter Donnelly, is already folding the model into its Mystra platform. DeepMind says it finds causal variants 30% of the time against 16% before, but weights are unreleased and commercial access is a private preview.
Coverage Details
Bias Distribution
- 50% of the sources lean Left, 50% of the sources are Center
Factuality
To view factuality data please Upgrade to Premium
