Backbone generation in protein design workflows

In this session, you’ll learn how to approach iterative design using modern computational tools. The lecture introduces backbone generation, a core concept in protein design, and explores how to evaluate designs using metrics like RMSD, diversity, and designability. We also cover common pipelines from RFdiffusion to ProteinMPNN to AlphaFold, along with strategies like binder design, motif scaffolding, and symmetric generation. Finally, the video provides a practical overview of RFdiffusion, including its inputs, outputs, limitations, and how it fits into a full design workflow. This is a high-level but practical guide to thinking about protein design pipelines and methodologies. More info about this Bootcamp can be found at https://rosettamlbootcamp2025.github.io/ Instructor: Ian Anderson Affiliation: UC Davis Chapters 00:00 – Capstone project overview and group assignments 01:44 – Project goals and design pipeline expectations 05:16 – What is backbone generation? 06:07 – Validating designs: RMSD and AlphaFold pipeline 07:20 – Diversity, novelty, and designability metrics 11:51 – Applications: binder design, scaffolding, symmetry 13:57 – RFdiffusion overview and methodology 19:01 – Inputs, outputs, limitations, and assignment setup