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 (blog)
15 months ago we released AlphaGenome, an AI model that can predict how genetic variants impact biological processes. It’s been great to see this community use AlphaGenome and we’ve learned a lot from the feedback and insights shared. Atlas seeks to address a few different pieces of feedback we’ve heard from y’all.
-
Precomputed scores that can be provided with higher quota. AlphaGenome is helpful for analyzing specific variants and has found widespread use in research, but we wanted to show researchers a big-picture view of variants across the entire genome.
-
A no-code solution for using AlphaGenome: The Atlas website provides an intuitive way to explore AlphaGenome scores for variants and regions. We’re also shipping a skill to accelerate larger workflows with Atlas using agentic tools like Antigravity.
-
A single score to help rapidly rank/ prioritise variants: The AlphaGenome Variant impact (AVI) score is a single standardised score that 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
Access the resources here:
Atlas has been the result of over a year of work and we are so excited to share this with you today and learn from your feedback. We’ll be running a workshop at ASHG in October so please stay tuned for those details.
Finally, I wanted to thank all our community collaborators and testers for the insights and feedback shared. We could not have done this without your input!
Amanda (on behalf of the AlphaGenome team)