Summer RosettaCon 2025 Workshops
1-day Protein Design Workshop
AI-Driven Design of Biologics: Hands-on Guide to Using State-of-the-Art ML Protein Models brought to you by the Rosetta Commons
One-day course surveying popular tools and workflow for modern protein design with brief exploration of common applications including antibody design and binder design.
Since 2021, artificial intelligence models have revolutionized AI-driven biologics development, enabling breakthroughs in structure prediction, sequence design, and protein engineering. This course equips researchers and professionals with the expertise to leverage cutting-edge tools for structure prediction (AlphaFold, ImmuneBuilder), protein engineering with protein language models (ESM, IgBERT, AntiBERTy) and structure based design (ProteinMPNN and RFDiffusion). Through a blend of lectures and hands-on exercises, participants will learn best practices for tool selection, method optimization, and design selection. By exploring real-world applications and emerging techniques, such as BindCraft and RFAntibody, attendees will gain a practical understanding of performance capabilities, limitations, and effective workflows.
Participants are expected to have some prior exposure to computational modeling tools (e.g. Python, R, COOT, Rosetta, AutoDock Vina, etc.) but limited experience applying them to their projects. They should be comfortable using Jupyter notebooks and prepared to explore topics such as evaluating metrics, determining appropriate sampling sizes, and selecting key adjustable parameters. While this course does not cover ligand docking or protein-protein docking, it is well-suited for those interested in antibody modeling and, potentially, enzyme design language models.
Topics to be covered:
- Building practical experience with AI-based modeling of proteins
- A breakdown of input formats, command lines, and analysis of output
- Hands-on exercises using real-world scenarios in antibody structure prediction, developability pre-screening, immunogen solubilization, and de novo binder design
- Discussion of and guidance on questions like how many models, in silico selection metrics and ranking, and how many to test in the lab
- Pipelining of protein design software and the critical use of an “oracle”
- Hardware, cloud computing, and SaaS for implementation of AI-powered protein design
Hands-on exercises in:
- structure prediction of proteins and antibodies
- application of protein language models to developability screening and enzyme engineering
- structure-based sequence design for immunogen solubilization
- structure-based backbone design
Application & Registration:
- Rosetta trainees: Lodging (but not other travel) costs will be covered by the Rosetta Commons.
- Industry participants: modest registration fee of $200 will support event sustainability.
Instructors:
David P. Nannemann, PhD, Vice President, Rosetta Commons Foundation
David is an expert in protein engineering and computational design, with extensive experience applying AI-driven modeling tools in an industry setting. He serves as Vice President of the Rosetta Commons Foundation and Industry Chair on the Rosetta Commons board, helping bridge academic advancements with industry applications. As Managing Member of Rosetta Design Group, he collaborates with companies of all sizes to tackle complex challenges in biologics design. David's deep expertise in leveraging cutting-edge tools like Rosetta, AlphaFold, and diffusion-based models for protein design make him an invaluable guide for participants looking to apply AI-driven biologics design in real-world settings.
Additional Local Instructors drawn from experts within the Rosetta Commons
This course also features expert instruction from senior graduate students and postdoctoral researchers trained in labs of the Rosetta Commons community. These instructors are advanced users who apply state-of-the-art computational modeling and design tools daily to solve cutting-edge problems in biologics research. Drawn from over 115 Rosetta Commons labs worldwide, they bring diverse perspectives and the latest advancements from the forefront of computational protein engineering.
Event dates
- Sunday 8/3 1pm - 6 pm
- Monday 8/4 8am - 2 pm
Event Location
Double Tree by Hilton Hotels Seattle Airport
How to train your model
An advanced protein design bootcamp/hackathon software developer event
This immersive, hands-on event is designed for researchers who are ready to move beyond “out-of-the-box” machine learning models for protein design. Participants will gain practical experience in training models tailored to their specific research needs, learning about the process, key variables, limitations, and possibilities of model development with a project-based learning experience.
Unlike traditional bootcamps, this program combines elements of both a bootcamp and hackathon—there is no formal instruction. Instead, participants will work in a high teacher-student ratio environment with a diverse team of experts providing guidance. Students must come prepared with a well-scoped project, a suitable dataset, and a solid understanding of existing models. This is not a novice-level course—prior ML experience is required. (For those new to ML, we offer an introductory bootcamp in the fall.)
A series of required preparatory meetings will help participants scope their project, prepare training/validation/testing data, and ensure readiness to dive in. The bootcamp bookends Summer RosettaCon (SRC), with model training development occurring on the Sunday & Monday before the conference, and results analysis, troubleshooting, and next-step planning happening the following Friday.
Participants are encouraged to install all necessary tools and resources on their lab or company clusters, but compute resources from Leviate Bio may be available as an alternative.
Application & Registration:
-
- Rosetta trainees: Lodging (but not other travel) costs will be covered by the Rosetta Commons.
- Industry participants: modest registration fee of $400 will support event sustainability.
- All participants must co-register for SRC.