Tuesday
Overview
Tuesday focuses on protein structure prediction using state-of-the-art tools. You’ll learn the theoretical foundations of protein folding, visualize structures with PyMOL, predict structures using AlphaFold2, and compare different prediction methods to understand their strengths and trade-offs.
By the end of Tuesday, you’ll be able to:
- Explain the protein folding problem, Anfinsen’s hypothesis, and Levinthal’s paradox in plain language.
- Navigate a protein structure in PyMOL: select residues, measure distances, and inspect interfaces.
- Describe how AlphaFold2’s MSA-based architecture (Evoformer + Structure Module) turns co-evolution into 3D coordinates.
- Interpret AlphaFold2 confidence metrics — pLDDT (per-residue) and PAE (inter-residue) — and know what each tells you.
- Choose between AlphaFold2 and ESMFold for a given task based on MSA availability, speed, and accuracy needs.
Modules
| Module | Topic | Description |
|---|---|---|
| 1. PyMOL and VSCode | Visualization & Development Tools | Learn protein structure visualization with PyMOL and remote development with VSCode |
| 2. Structure Prediction | Foundational Concepts | Understand the protein folding problem, Anfinsen’s hypothesis, Levinthal’s paradox, and structure prediction metrics |
| 3. AlphaFold2 | AlphaFold2 & OpenFold | Deep dive into AlphaFold2’s architecture, run ColabFold predictions, and interpret confidence scores |
| 4. ESMFold vs. AlphaFold2 | Comparing Prediction Methods | Compare ESMFold’s language model approach vs AF2’s MSA-based approach. Learn when to use each tool |