NVIDIA tutorial shows RFD4-Proteina protein design with reward-guided search
NVIDIA's technical blog published a tutorial on designing protein binders with RFD4-Proteina, an NVIDIA BioNeMo open model co-developed with the Institute for Protein Design at the University of Washington's School of Medicine. The model co-designs protein sequences and all-atom structures and uses reward-guided inference-time search to focus compute on the most promising candidates; the walkthrough targets PD-L1 binder candidates for computational biologists, ML engineers and drug discovery researchers. NVIDIA says users can take the open model for training, inference, method development and direct control over sampling code, or a NIM microservice with an HTTP interface, and that RFD4-Proteina is faster and uses less memory than RFD3, while the accompanying manuscript reports state-of-the-art results across a range of molecular tasks.
The post gives AI-biology teams a documented, end-to-end workflow — condition specs, open-model and NIM paths, and campaign and packaging skills for agents — for applying an open protein-design model to an oncology target, though the performance claims are NVIDIA's own.
Protein design is a search problem. Choose amino acids for an eight-residue protein sequence and more than 25 billion combinations are possible. Choose amino acids for a 100-residue protein sequence and the number of possibilities exceeds the number of atoms in the observable universe. Somewhere in that space is a protein that opens a path to a new therapy. There are too many combinations to search exhaustively, so finding it depends on knowing where to look.
RFD4-Proteina, an NVIDIA BioNeMo open model co-developed with the Institute for Protein Design at the University of Washington’s School of Medicine, navigates that space using patterns learned from nature. It co-designs protein sequences and all-atom structures, and uses reward-guided inference-time search to focus compute on the most promising candidates.
This tutorial shows computational biologists, ML engineers and drug discovery researchers how to design protein binder candidates against PD-L1, an oncology target, using reward-guided inference-time search and introduces experimental validation results across design tasks.
You can access RFD4-Proteina in two ways:
- Choose the open model for training, inference, method development, and direct control over sampling code or
- Choose the NIM microservice for rapid inference deployment with an HTTP interface
#NVIDIA#BioNeMo#RFD4-Proteina#Institute for Protein Design#University of Washington School of Medicine#NIM#RFD3#ProteinMPNN#Proteina-Complexa#OpenFold3#BioNeMo Agent Toolkit