AlphaGenome Atlas logo

Research / product dossier

AlphaGenome Atlas

Explore AI predictions for how 9B single-letter DNA changes may affect biology.

Product brief

What AlphaGenome Atlas does.

AlphaGenome Atlas is Google DeepMind's AI-powered map of how genetic mutations may affect human biology. Built by precomputing AlphaGenome predictions for all 9 billion possible single-letter DNA changes, the 1-petabyte dataset lets researchers explore and prioritize variants across both coding and non-coding regions. It's free to explore through a visual web interface, with API and Antigravity access for deeper research.

Why we selected it

A genuinely consequential research resource rather than a thin AI interface: a free visual explorer and API for precomputed predictions spanning all 9 billion single-letter human DNA changes. It brings a large-scale

Best for
Genomics and biomedical researchers
Category
Research
Daily picks
1
First selected
2026-09-09
AlphaGenome Atlas product preview

Product preview saved with our daily selection

Capability scan

What it can help with.

Only capabilities supported by the product information we collected are listed here.

01

Mutation-effect map

Provides an AI-powered map of how genetic mutations may affect human biology.

02

Precomputed variant predictions

Uses AlphaGenome predictions precomputed for all 9 billion possible single-letter DNA changes.

03

Coding and non-coding exploration

Lets researchers explore and prioritize variants across coding and non-coding regions.

04

Visual web interface

Offers a free visual web interface for exploration.

05

Research access options

Provides API and Antigravity access for deeper research.

Best-fit use cases

Explore possible effects of single-letter human DNA changes.
Prioritize variants across coding and non-coding regions.
Use the visual interface to investigate precomputed predictions.
Access the dataset through the API or Antigravity for deeper research.

FAQ

Before you open it.

What is AlphaGenome Atlas?

AlphaGenome Atlas is Google DeepMind's AI-powered map of how genetic mutations may affect human biology.

What variants does it cover?

It is built from predictions for all 9 billion possible single-letter DNA changes.

Can it be used for non-coding regions?

Yes. Researchers can explore and prioritize variants across both coding and non-coding regions.

How can researchers explore it?

It is free to explore through a visual web interface.

Are there options for deeper research access?

Yes. The description lists API and Antigravity access for deeper research.