DeepMind releases AlphaGenome Atlas covering 9 billion DNA variants

Google DeepMind has published precomputed predictions for every possible single-letter DNA change in the human genome, free for academic research.

Paylaş
DeepMind releases AlphaGenome Atlas covering 9 billion DNA variants

Google DeepMind has launched AlphaGenome Atlas, a platform holding predictions for the effects of 9 billion single-nucleotide variants, every single-letter change possible in the human genome. The company calls it the most comprehensive catalogue of how genetic mutations affect molecular biology, and it is available for academic research through a free website portal.

The platform builds on AlphaGenome, DeepMind's model for predicting how genetic variants affect biological processes. By precomputing those predictions at genome scale, the team turned a per-variant tool into a browsable resource. The dataset weighs 1 petabyte, more than 30 times the size of the AlphaFold Database.

A single score for coding and non-coding regions

Alongside the Atlas, DeepMind released the AlphaGenome Variant Impact (AVI) score, which condenses predictions from AlphaGenome and AlphaMissense, its model for protein-altering variants, into one number. The score works both for the 2% of the genome that codes for proteins and for the remaining 98% of non-coding sequence, which orchestrates gene activity and houses most trait-associated variants. DeepMind says its testing shows best-in-class performance across many variant pathogenicity and rare disease benchmarks. Feature attributions indicate which molecular processes, such as RNA splicing or gene expression, are predicted to be most disrupted by each variant.

Early use in rare disease work

Working with the GREGoR Consortium, Laura Covill and Anne O'Donnell-Luria of the Broad Institute and colleagues used the AVI score to prioritise variants overlooked in earlier research. They identified a variant affecting DNM1, a gene strongly linked to epileptic encephalopathy. The underlying AlphaGenome predictions showed the variant created an incorrect splice site, leading to an abnormal extension of the resulting protein. Experimental screens validated the prediction and turned up nearby variants with similar effects.

AlphaGenome Atlas is available from today through the website portal, the AlphaGenome API and as a skill in Google Antigravity. Full details are in DeepMind's announcement.

Devamını oku

Liquid AI تطلق أكبر نماذج LFM2.5 المفتوحة لتشغيل الوكلاء على الأجهزة

Liquid AI تطلق أكبر نماذج LFM2.5 المفتوحة لتشغيل الوكلاء على الأجهزة

أطلقت Liquid AI نموذج LFM2.5-24B-A2B كنموذج مفتوح، ويضم 24 مليار معامل إجمالا لكنه يشغل نحو 2.5 مليار منها في كل خطوة. وبحسب النتائج التي نشرتها الشركة، يولد النموذج المدرب للوكلاء الذين يستخدمون الأدوات 184 رمزا في الثانية على Apple M5 Max.

GlobalFeed Editor tarafından