The transcript traces the quest to understand how a protein’s amino‑acid sequence determines its three‑dimensional shape—a problem that has puzzled biologists for decades. Early work (Anfinsen, Kendrew) showed that the information for folding resides in the sequence itself, yet predicting the fold remained astronomically difficult. The biennial CASP challenge provided a benchmark for computational methods, and progress was slow until deep‑learning approaches emerged. DeepMind’s AlphaFold 2, trained on the Protein Data Bank and using transformer‑based Evoformer and Structure modules, achieved unprecedented accuracy in CASP 14 (scores ≥ 90), effectively solving the core protein‑folding problem. This breakthrough, together with RoseTTaFold and later AlphaFold 3, enabled rapid, high‑confidence structure predictions for hundreds of millions of proteins and opened avenues for AI‑driven protein design—creating novel enzymes, therapeutics, and sustainable materials. The achievement was recognized with the 2024 Nobel Prize in Chemistry awarded to David Baker, John Jumper, and Demis Hassabis, marking the start of a new era in computational biology where AI predicts not only static structures but also protein‑protein and protein‑small‑molecule interactions, while challenges remain in modeling the full cellular context of protein function.
1. Proteins are microscopic molecular machines essential to life.
2. Proteins first appeared at least 3.7 billion years ago.
3. Proteins are built from 20 different types of amino acids connected in polypeptide chains.
4. A protein's specific function is determined by its three-dimensional folded shape.
5. The sequence of amino acids for a protein is encoded in a cell's DNA.
6. The "protein folding problem" involves predicting a protein's 3D structure from its amino acid sequence.
7. X-ray crystallography is a technique used to reveal atomic protein structures.
8. The Protein Data Bank (PDB) contains structural data for more than 200,000 proteins.
9. Christian Anfinsen discovered that a protein's 3D structure is encoded in its amino acid sequence.
10. Proteins form secondary structures like alpha helices and beta sheets.
11. CASP (Critical Assessment of Structure Prediction) is a biannual challenge that compares computational predictions to experimental structures.
12. DeepMind developed AlphaFold, an artificial intelligence system for protein structure prediction.
13. AlphaFold 2 achieved high accuracy in the CASP 14 challenge.
14. In July 2022, DeepMind released structure predictions for 218 million proteins.
15. RFdiffusion is a generative AI system used to design new protein backbones.
16. RoseTTAFold All-Atom predicts the structures of protein assemblies and small molecules.
17. AlphaFold 3 uses a diffusion-based method to predict interactions between proteins and other molecules.
18. David Baker, John Jumper, and Demis Hassabis won the 2024 Nobel Prize in Chemistry for their work on protein structure prediction and design.