DeepMind's AlphaGenome Atlas precomputes effects of ~9 billion single-letter DNA mutations
Tkins · reddit · 2026-09-08
Google DeepMind launched AlphaGenome Atlas, a searchable 1-petabyte database precomputing AlphaGenome predictions for roughly 9 billion possible single-nucleotide mutations across the human genome — turning the model into a "genomic search engine."
- Non-coding DNA is the big win: the 98% of the genome that doesn't code for proteins is hardest to interpret, and the Atlas helps flag which mutations there may matter for disease.
- New AVI score: the AlphaGenome Variant Impact score condenses many predicted molecular effects into one number for ranking large variant sets.
- Early use cases: predictions helped identify a likely disease-causing mutation in a rare-disease case, and another analysis found new associations between non-coding variants and traits like BMI.
- Caveats: these are AI predictions of molecular effects, not experimental proof of disease causation; key findings still need lab and clinical validation.
More from Research
- AutoResearch agent produces weird 'forking' loss curves, and the team dug into the science — ZimingLiu11 · 2026-10-27
- No π0 weights changed: Motoniq layer lifts real-robot precision fitting from 0/10 to 9/10 — hbouammar · 2026-10-27
- JockeyBench uses Japanese horse racing to test LLM agents — all betting strategies lose money — Zhongwei Xie · 2026-10-27
- Stanford's PaperFormalizer turns research papers into Lean-verified versions with author in the loop — StanfordUniversity · 2026-10-27
- Cohere's Invent-A-Dataset paper: zero-seed synthetic data beats frontier APIs on quality and diversity — idanbeck · 2026-10-27
- Hugging Face TRL makes score centering + masked importance sampling default in async GRPO — _lewtun · 2026-10-27