# AlphaGenome

Source: https://onco.cc/technologies/alphagenome/  
OnCo record `alphagenome` (Technology). Data CC BY-NC 4.0, attribute "Data from OnCo (onco.cc)"; commercial use needs a licence.

## TL;DR

Reads a million letters of DNA at once and predicts how a mutation changes gene regulation, splicing and chromatin.

## Summary

AlphaGenome is Google DeepMind's long-context sequence model with multi-task regulatory heads, which reads up to a million bases of DNA and predicts, at single-base resolution, how variants change gene expression, splicing, chromatin accessibility and other regulatory signals in one model. It launched as a preview API in June 2025, and the Nature paper followed in January 2026. By unifying tasks that previously needed separate models it is relevant to interpreting non-coding cancer drivers in enhancers, promoters and splice sites. It is for research use, and cancer-specific validation is ongoing, so its predictions remain hypotheses to test rather than clinical evidence. For a newcomer: AlphaGenome reads long stretches of DNA and predicts how a mutation outside a gene would change the gene's behaviour.

## Fields

- Kind: Technology
- Status: emerging
- Last checked: 2026-09-08
- Tags: foundation-model; genome
- Principle: Long-context sequence model with multi-task regulatory heads.
- Since: 2025
- Strengths: Breadth of predicted modalities
- Limitations: Research use; cancer-specific validation ongoing

## Sources

- DeepMind AlphaGenome: https://deepmind.google/discover/blog/alphagenome-ai-for-better-understanding-the-genome/

## Connected records

- fronts: [AI & Computation](https://onco.cc/fronts/ai-computation/)
- companies: [Google DeepMind (and Google Research)](https://onco.cc/companies/google-deepmind/)

---
JSON: https://onco.cc/api/v1/entities/alphagenome.json