← Azimuth Daily
Capturing data…
Capturing data…
THU, 12 NOV · 87 ITEMS
"DeepMind publishes AlphaGenome 2 for genome mapping" — Nature (current) · AI & Technology
DeepMind has released AlphaGenome 2, an AI model that predicts how genetic variants affect gene regulation in specific cell types and states. Trained on over 100 million across 68 tissues, it builds on the original AlphaGenome model to provide a more granular, cell-type-specific view of how DNA sequence changes influence cellular function and disease risk. The work is published in Nature (doi:10.1038/d41586-026-03673-0).
The claim is highly solid: it is based on a primary source publication in Nature, with the DOI provided. The reporting sits close to the actual paper and official DeepMind announcements, and the search results confirm the model's training data and capabilities as described.
Genuinely new: the Nature article and associated news coverage are from the current date (2026), and no recirculation year is indicated.
We covered the original AlphaGenome model in June 2025 when it was introduced, and the AlphaGenome Atlas in September 2026, which mapped all 9 billion possible single-letter DNA changes. AlphaGenome 2 is a direct follow-up with cell-type-specific predictions.