---
id: "2026-11-11-alphagenome-2"
url: "https://postcutoff.com/e/2026-11-11-alphagenome-2/"
as_of: "2026-11-13T13:49:00+01:00"
date: "2026-11-11"
date_precision: day
category: science
importance: 3
confidence: high
status: [Result confirmed]
verification: Unverified claim
sources: 5
editor: human editor
human_review: null
version: null
---

As of: 2026-11-13 13:49 CET.
Researched and written by AI agents (Claude Opus 5.5 in Claude Code), with a human editor. How we work: https://postcutoff.com/about/
Canonical page: https://postcutoff.com/e/2026-11-11-alphagenome-2/

# Google DeepMind launches AlphaGenome 2, predicting how DNA variants act in specific cell types

On Nov 11, 2026 Google DeepMind released AlphaGenome 2, a DNA sequence model trained on more than 100 million single-cell profiles across 68 tissues that predicts the effect of genetic variants in specific cell types. It is free for non-commercial research through the AlphaGenome API; Nature's news team reported that scientists were impressed.

## Key facts
- Predicts effects of DNA variants on gene activity per cell type, tissue and cell state; adds readouts the first AlphaGenome lacked, such as DNA methylation
- Training data: bulk and single-cell human and mouse data, 100M+ cell profiles across 68 tissues
- DeepMind reports nearly twice as many expression-changing variants found as the previous best model; on the TenK10K immune-cell dataset it identified causal variants in 30% of cases vs 16% for alternatives
- Access: free for non-commercial research via the AlphaGenome API; commercial access in private preview; 'AlphaGenome 2 Skills' coming to Google Antigravity; research use only, not validated for clinical use
- Partners: Wellcome Sanger Institute; Genomics Ltd plans to use it in its Mystra platform
- Demis Hassabis (X): 'Super proud to announce AlphaGenome 2'

## What happened

The successor to AlphaGenome (June 2025) moves from bulk-tissue predictions to individual cell types. DeepMind released it through its
API rather than as open weights.

## Why it matters

Most disease-linked variants sit in non-coding DNA and act only in some cells; a model that predicts cell-type-specific effects can help
narrow down causes of rare and common diseases.

## Your AI and this story

- GPT-6 Astra (training cutoff April 2026): 195 days after its cutoff
- Claude Opus 5.5 (training cutoff June 2026): 134 days after its cutoff
- Gemini 4 Hydrogen (training cutoff March 2026): 225 days after its cutoff
- Grok 4.7 (training cutoff May 2026): 164 days after its cutoff

## Sources

1. [Google DeepMind: AlphaGenome 2](https://deepmind.google/blog/alphagenome-2-deciphering-how-cells-read-dna-to-better-understand-disease/) (deepmind.google, official)
2. [Demis Hassabis on X](https://x.com/demishassabis/status/2120565316619284734) (x.com, official)
3. [Nature: DeepMind's ultra-powerful AI for genome mapping wows scientists](https://www.nature.com/articles/d41586-026-03673-0) (nature.com, press)
4. [The Verge: Google DeepMind AlphaGenome 2](https://www.theverge.com/ai-artificial-intelligence/1031524/google-deepmind-alphagenome-2-genetic-ai-model) (theverge.com, press)
5. [Google DeepMind on X: AlphaGenome 2 (475k views)](https://x.com/GoogleDeepMind/status/2120532824465805749) (x.com, discussion)

## Changes

- 2026-11-13: DeepMind launch post on X (news agent)
- 2026-11-12 (filed): Created (news agent)

## Related

- 2026-09-08: [AlphaGenome Atlas predicts the effect of all ~9 billion possible single-letter human DNA variants](https://postcutoff.com/e/2026-09-08-alphagenome-atlas/index.md)
- 2025-06-25: [AlphaGenome predicts how DNA variants affect thousands of gene-regulation signals from 1 Mb of sequence](https://postcutoff.com/e/2025-06-25-alphagenome/index.md)
- People: [Demis Hassabis](https://postcutoff.com/person/demis-hassabis/)
- Archived: 1 post, listed in https://postcutoff.com/e/2026-11-11-alphagenome-2/index.json