Google DeepMind Maps All 9 Billion Possible Human Gene Mutations in New AlphaGenome Atlas
13 Sources
[1]
DeepMind's new genome 'atlas' charts effects of all 9 billion human gene mutations
The human genome is an easy place to get lost. Only 2% of its 3 billion letters encode proteins and the rest is diabolically hard to decipher. An AI-generated 'atlas' of the human genome unveiled today by Google DeepMind aims to guide scientists through our biological code. One of the most common
[2]
Google's AI genome system evaluates every possible one-base change
On Tuesday, Google announced AlphaGenome Atlas, a resource that attempts to predict the consequences of every possible single-base variant in the human genome. The human genome is about 3 billion bases long, so trying the other three DNA bases that don't appear in our reference genome means sending
[3]
DeepMind's AI Just Mapped 9 Billion Possible DNA Variants
Greg Uyeno is a freelance science journalist based in New York City. DNA is often explained as a codebook or set of instructions for producing proteins, and ultimately, life. Some stretches of DNA, called genes, code for proteins, but the vast majority of DNA is considered "non-coding." Some of it
[4]
New Google DeepMind atlas could transform our understanding of genetic diseases
I agree my information will be processed in accordance with the Scientific American and Springer Nature Limited Privacy Policy. We leverage third party services to both verify and deliver email. By providing your email address, you also consent to having the email address shared with third parties
[5]
Google's Atlas of the human genome could pave the way for new treatments
Google DeepMind has unveiled an AI tool that its scientists claim could help unravel the mysteries of the human genome and transform our understanding of biology, accelerating scientific research and ultimately paving the way for new treatments for diseases. The platform, called AlphaGenome Atlas,
[6]
Google DeepMind rises above the AI scrum with genome atlas
While OpenAI and Anthropic fight over whose models can escape their sandbox more alarmingly, Google's DeepMind team has once again shown how machine learning can also be used to advance science for humanity's benefit. On Tuesday, the Chocolate Factory's crack team of AI researchers unveiled
[7]
AlphaGenome Atlas maps billions of genetic changes with AI
Stowers Institute for Medical ResearchSep 9 2026Reviewed The human genome contains approximately 3 billion DNA letters, creating more than 9 billion possible single-letter changes. Testing the effects of each change in a laboratory would be practically impossible. Google DeepMind's new AlphaGenome
[8]
Google's map of every possible DNA typo could speed up rare disease research
The Atlas is free for academic use and heading to Google Cloud for everyone else, which is exactly the boundary the AI Act's research exemption turns on DeepMind has released AlphaGenome Atlas, a petabyte of precomputed predictions for all 9 billion possible single-letter changes in the human
[9]
AlphaGenome Atlas: Molecular predictions for 9 Billion human DNA variants
Our testing shows that the AVI score provides best-in-class performance across many variant pathogenicity and rare disease benchmarks. To help interpret these scores, we also calculated AVI feature attributions that highlight which molecular processes -- like RNA splicing or gene expression -- are
[10]
Google DeepMind publishes AI-powered predictions for the effect of all 9 billion possible single-point mutations to human DNA | Fortune
Google DeepMind said Tuesday that it has used artificial intelligence to predict the biological consequences of all 9 billion possible single-letter changes to human DNA, and is making the resulting database available free to academic researchers worldwide. AlphaGenome Atlas, as DeepMind calls the
[11]
AlphaGenome Atlas: a high-resolution map of human DNA
The human genome is made of about 3 billion base pairs of DNA -- but much of it remains a mystery. Scientists understand the 2% of the human genome that codes for proteins relatively well, but have only limited knowledge of the remaining 98%. Our AlphaGenome model has already shown how single
[12]
Google DeepMind's AlphaGenome Atlas maps all 9 billion possible human DNA changes
Google DeepMind's AlphaGenome Atlas maps all 9 billion possible human DNA changes Google LLC's DeepMind research unit said today it has used its AlphaGenome artificial intelligence model to predict the biological consequences of the more than nine billion possible single-letter changes to human
[13]
Google unveils AlphaGenome Atlas mapping DNA mutation effects By Investing.com
Investing.com - Google released AlphaGenome Atlas on Tuesday, a database that predicts the effects of every possible single nucleotide variant in the human genome. The company used its AlphaGenome AI model to pre-calculate the regulatory impact of all 9 billion single-letter genetic changes,
Share
Copy Link
Google DeepMind unveiled the AlphaGenome Atlas, an AI-generated atlas of the human genome with predictions for all 9 billion possible single DNA letter mutations. The 1-petabyte database includes the AlphaGenome Variant Impact (AVI) score to help researchers identify disease-causing mutations and decode non-coding DNA's regulatory functions.
Google DeepMind Releases Comprehensive AI-Generated Atlas of the Human Genome
Google DeepMind unveiled the AlphaGenome Atlas on September 8, a groundbreaking AI-generated atlas of the human genome that charts predictions for all 9 billion possible single DNA letter mutations
1
3
. The database builds on the AlphaGenome AI model released last year and represents the most comprehensive catalogue of how gene mutations affect molecular biology5
. The human genome contains roughly 3 billion base pairs, and at each position there are three possible single-nucleotide substitutions, creating the 9 billion variants now catalogued in the Atlas3
. The complete petabyte dataset is freely available for non-commercial use, with commercial licensing available through Google Cloud4
5
.Decoding Non-Coding DNA and Its Role in Genetic Disease Research
The AlphaGenome Atlas focuses primarily on non-coding DNA, which comprises 98 percent of the human genome but doesn't directly code for proteins
4
. This non-coding DNA plays a critical role in regulating gene expression, controlling when and where genes are activated, and determining chromatin structure2
. Understanding how single DNA letter mutations in these regulatory regions affect biology is fundamental to understanding most diseases, according to Carl de Boer, a genomicist at the University of British Columbia3
. The Atlas provides thousands of predictions about different potential effects of each variant, from how it affects gene expression in specific tissues to transcription factor binding and splice site usage1
2
.AlphaGenome Variant Impact Score Simplifies Complex Genomic Predictions
A key feature of the AlphaGenome Atlas is the AlphaGenome Variant Impact (AVI) score, a single-number metric that helps researchers quickly assess whether a variant warrants deeper investigation
1
4
. This was a repeated request from the approximately 9,000 researchers who had accessed AlphaGenome's predictions through its automated programming interface since its release1
. The AVI score combines AlphaGenome and AlphaMissense predictions and reliably distinguished disease-causing mutations from harmless changes in clinical genomics databases, according to a preprint released by DeepMind and academic researchers1
. The score helped a team at the Broad Institute in Cambridge, Massachusetts, prioritize a non-coding variant as a possible cause of severe epilepsy1
.Computational Advancements Enable Massive Scale Analysis
Creating the AlphaGenome Atlas required significant computational advancements. When DeepMind started the project, early estimates indicated they would need to improve calculation speed by a factor of 80 to compile the Atlas in reasonable time, according to Žiga Avsec, genomics lead at Google DeepMind
3
. The team achieved this through model distillation, GPU kernel optimization, and elimination of redundant calculations3
. The effort was inspired by DeepMind's AlphaFold database of over 200 million protein-structure predictions, which has been accessed by millions of users and helped Demis Hassabis win a share of the 2024 Nobel Prize in Chemistry1
4
. The AlphaGenome Atlas is 30 times larger than AlphaFold's database, containing approximately 1 petabyte of data4
.Accelerating Rare Disease Diagnosis and Drug Discovery
The predictive map of DNA letter changes could accelerate rare disease diagnosis and drug discovery by helping researchers identify disease-causing mutations without extensive computational work
1
4
. Rare-disease researchers have typically relied on less computationally demanding models because applying tools like AlphaGenome across an entire human genome was unfeasible for most researchers, according to computational biologists Mafalda Dias and Jonathan Frazer at the Centre for Genomic Regulation in Barcelona1
. Jonathan Sebat, a psychiatric geneticist at the University of California, San Diego, noted that workflows in labs can be streamlined significantly because researchers can simply look up genomic predictions rather than computing them4
. The Atlas removes computational barriers and provides instant access through a web portal, as well as integration with Google's Antigravity agentic development platform5
.Related Stories
Uncovering Hidden Rules of DNA Regulatory Elements
Beyond disease diagnosis, the AlphaGenome Atlas aims to reveal hidden rules governing how DNA sequences control gene activity through regulatory elements called motifs
1
. Using atlas predictions as a guide, researchers mapped thousands of DNA motifs across the human genome and inferred their roles in different cell types, whether activating genes, repressing them, or altering DNA accessibility1
. Julia Zeitlinger, a molecular biologist at the Stowers Institute for Medical Research in Kansas City, Missouri, described the work as providing "a searchable dictionary for non-coding DNA"1
. This capability matters because regulatory elements can interact in complicated ways, with effects varying across different cells and tissues, and some influencing genes located far away in the genome3
.Limitations and Need for Experimental Validation
While the AlphaGenome Atlas represents a significant advance, it has important limitations that require experimental validation. The model examines 1 million base pairs surrounding each variant, but some DNA sequences called enhancers can regulate genes over very long distances beyond the model's field of view
3
. Many diseases involve multiple genetic variants rather than single changes, adding complexity4
. Martin Kircher, a bioinformatician at the Max Delbrück Centre for Molecular Medicine in Berlin, emphasized that the Atlas won't replace experiments or detailed case analysis in disease diagnosis1
. The Atlas is also less accurate than AlphaFold and meant to serve as a starting point for research questions involving variants across the genome4
. AlphaGenome was trained only on limited cell types that biologists have studied exhaustively and is currently limited to sequences from mice and humans2
.References
Summarized by
Navi
[4]
Related Stories
Google DeepMind's AlphaGenome AI tackles the 98% of human genome that doesn't code for proteins
28 Jan 2026•Science and Research
DeepMind's AlphaGenome: Decoding the 'Dark Matter' of DNA with AI
26 Jun 2025•Science and Research
Evo-2: AI Model Revolutionizes Genomic Research and DNA Generation
20 Feb 2025•Science and Research
Recent Highlights
1
Meta Unveils Camera-Free AI Glasses and Muse Integration Amid Privacy Debate at Connect 2026
Technology
2
OpenAI Halts AI Agent Training After Rogue Systems Breach Government Websites Across Multiple Countries
Technology
3
Bill Gates Warns AI Could Cause a Billion Deaths, Calls for Mandatory Government Regulation
Policy and Regulation
Recent Highlights
Today's Top Stories
Don’t drown in AI news. We cut through the noise - filtering, ranking and summarizing the most important AI news, breakthroughs and research daily. Follow topics that matter to you and stay ahead.
