Building a Multi-Modal Cancer Variant Classifier with Structural Validation | New York City .

Members-Only

Recent Talks & Demos are for members only

Exclusive feed

You must be an AI Tinkerers active member to view these talks and demos.

November 17, 2025 · New York City

Evo 2 AlphaFold3 Cancer Classifier

Chaining Evo 2 and AlphaFold3 to classify cancer variants, this talk details a pipeline producing tiered therapeutic guidance validated with 3D protein-drug interactions.

Video
Overview
Links
Tech stack
  • Evo2
    Evo2 is a 7-billion parameter genomic foundation model trained on 800 billion tokens to predict and design DNA, RNA, and protein sequences at single-nucleotide resolution.
    Developed by the Arc Institute, Evo2 leverages a StripedHyena architecture to process biological sequences across massive scales (up to 1-megabase contexts). It outperforms previous models by treating the genome as a unified language, enabling zero-shot prediction of mutation effects and the generative design of complex regulatory elements. By training on the 800-billion-token OpenGenome dataset, Evo2 captures multi-scale biological structures from molecular folding to whole-genome organization.
  • AlphaFold3
    Google DeepMind’s latest model predicts the structure and interactions of all life’s molecules with unprecedented accuracy.
    AlphaFold3 moves beyond proteins to model DNA, RNA, and ligands in a single unified system. By utilizing a diffusion-based architecture, it predicts complex molecular interactions (such as a drug molecule binding to a specific protein receptor) with a 50% improvement over existing specialized tools. This capability allows researchers at institutions like the Francis Crick Institute to accelerate drug discovery and genomic research by visualizing how biological machinery functions at the atomic level.

Related projects