Intro to Deep Learning in Biology
Computational Biology @ Berkeley and Machine Learning @ Berkeley invite you to our joint Deep Learning for Biology workshop!
Ever wonder how deep learning is transforming biology — from mapping the 3D structures of proteins to designing new genes and therapies?
Join us for an exploration into how AI models are revolutionizing prediction, generation, and discovery across sequence, structure, and cellular systems.
📍Exact room location will be announced via slack and email on the day of the event!
We’ll cover:
CryoEM → Structure Prediction
• How cryo-electron microscopy and AlphaFold 2 enable atomic-level understanding of biomolecules.
• Deep geometric networks that predict function from structure.
Synthetic Biology → Generative Models
• AI for designing new proteins and genetic sequences.
• Inside RFdiffusion and Evo2, models that learn to create new biological organisms.
Interaction Prediction → Molecular Communication
• Predicting how proteins, ligands, and nucleic acids interact.
• Featuring Boltz, presented by Mahesh, showing how AI can model physical forces between molecules.
Gene Therapies & Genotype–Phenotype Modeling → Virtual Cells
• Using State, GEARS, and VEP (via Evo2) to simulate cell-level behavior and test gene-therapy strategies in silico.
Foundation Models → Universal Biological Intelligence
• From ESM to scGPT, exploring foundation models trained across proteins, genomes, and single-cell data.
• How cross-modal learning is unifying biology into a shared computational framework.
Why it matters:
Deep learning is redefining how we understand, predict, and engineer life — enabling smarter drug design, personalized medicine, and AI-driven biological discovery at every scale.