RFdiffusion
Baker Lab / IPD (University of Washington)
Diffusion-based generative model for de novo protein backbone design. Generates novel protein structures conditioned on binding targets, symmetry, or functional sites.
Best For
De novo binder design, enzyme design, symmetric oligomer design
License
Open Source (MIT)
Strengths
- +MIT license
- +Versatile conditioning
- +Strong experimental validation
Limitations
- −Generates backbones only (needs ProteinMPNN for sequences)
- −Success rates target-dependent
R&D Pipeline Coverage
Related Tools
ProteinMPNN
Baker Lab / IPD (University of Washington)
Inverse folding model: generates amino acid sequences predicted to fold into a target 3D backbone structure. Standard component of all modern protein design pipelines.
RFdiffusion2
Baker Lab / IPD (University of Washington)
Successor to RFdiffusion using flow matching. Designs enzymes directly from active site geometry (theozyme) specifications.
BindCraft
EPFL / MIT (Pacesa, Ovchinnikov, Correia)
One-shot automated pipeline for de novo protein binder design. Backpropagates through AlphaFold2 to hallucinate binders. 10-100% experimental success rates.
More in Generative Design
RFdiffusion2
Baker Lab / IPD (University of Washington)
Successor to RFdiffusion using flow matching. Designs enzymes directly from active site geometry (theozyme) specifications.
ProteinMPNN
Baker Lab / IPD (University of Washington)
Inverse folding model: generates amino acid sequences predicted to fold into a target 3D backbone structure. Standard component of all modern protein design pipelines.
LigandMPNN
Baker Lab / IPD (University of Washington)
Extension of ProteinMPNN that conditions sequence design on bound ligands, small molecules, metals, and nucleotides.
Stay updated on RFdiffusion
Weekly newsletter covering AI tool releases, benchmarks, and what practitioners actually use.