The UW Medicine Institute for Protein Design and Nvidia have released RFD4-Proteina, an open-source AI protein design model that outperforms baseline benchmarks in computational design success and generation throughput. Developed using a year-long Nvidia compute grant funded by Jensen and Lori Huang, the model enables rapid enzyme design for therapeutic and environmental applications. Nvidia simultaneously announced expanded healthcare AI partnerships with Siemens Healthineers and Danaher.
RFD4-Proteina development and distribution
- ▪The UW Medicine Institute for Protein Design and Nvidia released RFD4-Proteina, an open-source AI protein design model that combines programmable protein design with search capabilities.
- ▪RFD4-Proteina is available through partners including Amazon Bio Discovery, Anthropic, Benchling, Databricks, Google Cloud, Hugging Face, Nebius, OpenAI, and Tamarind Bio.
- ▪RFD4-Proteina was co-developed by UW doctoral students Nathaniel Corley, Maxwell Kazman, and Rafi Brent, alongside visiting University of Oxford student Kieran Didi.
RFD4-Proteina capabilities
- ▪According to a research draft, RFD4-Proteina achieved higher computational design success rates and generation throughput than evaluated baselines across four benchmark suites: protein binding, small-molecule binding, enzyme scaffolding, and DNA binding.
- ▪The RFD4-Proteina model co-developed by Nvidia and the University of Washington generates both protein sequences and three-dimensional structures.
Seth Woodbury's research
- ▪UW predoctoral candidate Seth Woodbury used RFD4-Proteina to design a pesticide-degrading enzyme in a few days that performed multiple times better than his previous best design.
- ▪University of Washington bioengineering predoctoral candidate Seth Woodbury published a paper on enzyme research in the journal Nature in 2025.
Healthcare AI collaborations
- ▪At GTC Berlin in October 2026, Nvidia announced that companies including Biolevate, Boltz, and Automata are using its BioNeMo platform to develop scientific agents connecting drug design and lab workflows.
- ▪Hologen is combining its models with Nvidia Nemotron to build AI clinical scientists for analyzing imaging and clinical data.
- ▪At GTC Berlin in October 2026, Nvidia announced healthcare AI collaborations, including Siemens Healthineers using Nemotron models for adaptive radiotherapy co-planning and Danaher incorporating RFD4-Proteina in its protein design foundry.
Nvidia computing grants to UW
- ▪The UW Institute for Protein Design began accessing Nvidia graphics processing units through cloud provider CoreWeave in June 2026 under a grant expected to last one year.
- ▪In October 2026, the UW Institute for Protein Design announced receiving a computing grant from Nvidia CEO Jensen Huang and Lori Huang through their foundation.
Story comments
Loading comments…