AlphaGenome Atlas: A predictive map of every possible DNA letter change in the human genome

[hotblack]

Well this looks interesting and worthy of its own thread (although feel free to move elsewhere)

Today, we are introducing AlphaGenome Atlas: a platform containing predictions for the effects of 9 billion single-nucleotide variants — every single-letter change possible — in the human genome. It is the most comprehensive catalogue of how genetic mutations affect molecular biology, and it is available for academic research through an intuitive and free-to-use website portal.

Announcement

Here’s the website itself and blurb

AlphaGenome Atlas is an integrated data resource that predicts the functional impact of all 9 billion possible single nucleotide variants (SNVs) across the human genome. By unifying coding and non-coding predictive models, it streamlines the prioritisation and interpretation of variants.

• Unified Variant impact scoring: Combines coding and non-coding predictions into a single, standardized AlphaGenome Variant Impact (AVI) score.
• Genome-wide scale: Access precomputed variant effect predictions spanning the entire human genome.
• Zero-code exploration: Spot-check individual variants and dive into granular genomic context directly in your browser.
• Seamless agentic integration: Connect effortlessly with AI agent workflows to scale analyses and transition to Atlas website visualisations.

 

[hotblack]

There is a paper here too

AlphaGenome Atlas: in silico mutagenesis of the entire human genome improves prioritization and interpretation of non-coding variants, 2026, Cheng, et. al.

Abstract
A major challenge in genomics is deciphering the functional consequences of non-coding genetic variation. Here we present AlphaGenome Atlas, a comprehensive resource that enables the joint interpretation and prioritization of variant effects across the entire human genome.

Using AlphaGenome, we predicted the regulatory effects across thousands of molecular phenotypes for every possible human single nucleotide variant and many observed indels. These predictions were then used to derive a unified and interpretable AlphaGenome Variant Impact (AVI) score and to map cis-regulatory motifs across the genome. AVI achieved state-of-the-art performance across diverse benchmarks with improved prioritization of deleterious non- coding variants.

Application of the combined Atlas resource helped solve an epileptic encephalopathy rare disease case, increased the statistical power to detect rare non-coding variants driving population-level phenotypes, and enhanced the mechanistic interpretation of these variants. Thus, AlphaGenome Atlas improves the prioritization and molecular interpretation of non-coding variants with genetic and clinical significance.

PDF

 

[hotblack]

AlphaGenome Atlas is a massive 1-petabyte dataset, more than 30 times larger than the AlphaFold Database.

To help scientists quickly find the most impactful genetic changes, we are also releasing the AlphaGenome Variant Impact (AVI) score. The AVI combines the strengths of AlphaGenome and AlphaMissense — our model for predicting the impact of protein-altering DNA variants — condensing both models’ predictions into a single number. Now, researchers can rapidly rank variants and interpret their molecular effects at the same time.

Our trusted external collaborators have already used AlphaGenome Atlas to identify and experimentally verify key variants in unsolved rare disease research and find rare variants associated with common traits.

So next up is for us to run all the variants from DecodeME through this and find which ones have high AVI scores and where right? Then we can do the same with SequenceME in the years to come…

More articles on this from Nature and Scientific American

 

[forestglip]

Our existing AlphaGenome thread: Advancing regulatory variant effect prediction with AlphaGenome 2026 Avsec et al

It looks like this is an interactive version of what previously required the API.

 

[Creekside]

That's fine for problems caused by a single genetic variation. Wouldn't there also be lots of diseases that involve multiple variants or simply dependencies on other genes? Maybe lx123 causes problems if the individual also has ry678 and doesn't have kj456 or sd987. Also, would epigenetic variations play a role for applying this tool for individuals?

Not dismissing the tool's value. Just pointing out that there's a lot more to diseases than variations in single genes.

 

[hotblack]

Our existing AlphaGenome thread: Advancing regulatory variant effect prediction with AlphaGenome 2026 Avsec et al

It looks like this is an interactive version of what previously required the API.

Yes, with precomputed results from AlphaGenome and Missense.

 

[hotblack]

Not dismissing the tool's value. Just pointing out that there's a lot more to diseases than variations in single genes.

Of course. But the more tools we have the better and being able to get an understanding of where to start looking in a pile of variants and an idea of what those variants may do seems hugely useful.

Does this variant change expression or rna splicing? What tissues is this relevant in? Etc etc. these are things we already have some databases for and these are only predictions so findings will need verifying but I can’t wait to see how people smarter than me use it.