ProteinMPNN
Fast, robust inverse folding — sequences for a target backbone
ProteinMPNN solves the inverse folding problem: given a protein backbone, it predicts amino-acid sequences that are likely to fold into it. It is the standard sequence-design partner for RFdiffusion backbones.
Highlights
- Dramatically higher experimental success rates than physics-based design.
- Handles monomers, complexes and symmetric assemblies, with position-wise constraints.
- MIT licensed and extremely fast — thousands of sequences per minute on a GPU.
Related tools
Open source
AI Models
RFdiffusion
Generative diffusion for de novo protein backbone design
Python libraryCommand lineModel weights
Open source
AI Models
BindCraft
One-shot de novo binder design with AlphaFold2 backpropagation
Python libraryCommand line
Open source
AI Models
ESM3 & ESM C
Frontier protein language models for representation and design
Python libraryREST APIModel weights