What is the difference between RFdiffusion and ProteinMPNN?

RFdiffusion designs the 3D coordinates of the protein backbone (the global fold), while ProteinMPNN is used afterward to determine the specific amino acid sequence that will fold into that shape.

Does this skill support symmetric proteins?

Yes, it supports multiple symmetry types including cyclic (C2-C6), dihedral (D2-D4), and higher-order symmetries like tetrahedral and octahedral.

Do I need a GPU to run this skill?

Yes, RFdiffusion requires a CUDA-enabled GPU with at least 16GB of VRAM (24GB recommended) to perform the diffusion simulations efficiently.

What is RFdiffusion used for?

RFdiffusion is used to generate the 3D backbone structures (the 'shape') of proteins, which is the first step in designing new binders, novel monomers, or symmetric complexes.

Can I design binders for a specific target with this skill?

Yes, you can provide a target PDB file and specify 'hotspot' residues to guide the model to generate binders that interact with specific regions of your target protein.

RFdiffusion Protein Design

Name: RFdiffusion Protein Design
Author: adaptyvbio

byadaptyvbio

•

データサイエンスとML

Generates novel protein backbones and binder scaffolds using diffusion-based generative modeling for de novo protein structure design.

概要

RFdiffusion is a specialized skill for Claude Code that enables researchers and bio-engineers to design protein structures from scratch or around specific target motifs. It leverages a powerful diffusion model to create high-quality protein backbones, supporting the creation of binders, symmetric oligomers, and scaffolded functional motifs. This skill automates the configuration of complex parameters like contig maps and hotspot residues, facilitating a streamlined workflow from initial target identification to validated backbone generation within a computational biology pipeline.

主な機能

Support for symmetric oligomer assembly (C, D, and polyhedral symmetries)
De novo protein backbone generation from scratch
Binder scaffold generation for specific protein targets
Partial diffusion for structural refinement and existing backbone modification
75 GitHub stars
Functional motif scaffolding with coordinate preservation

ユースケース

Creating novel enzyme scaffolds by grafting active sites onto new backbones
Generating symmetric protein assemblies for nanotechnology or vaccine design
Designing high-affinity protein binders for therapeutic targets

概要

主な機能

Support for symmetric oligomer assembly (C, D, and polyhedral symmetries)
De novo protein backbone generation from scratch
Binder scaffold generation for specific protein targets
Partial diffusion for structural refinement and existing backbone modification
75 GitHub stars
Functional motif scaffolding with coordinate preservation

ユースケース

Creating novel enzyme scaffolds by grafting active sites onto new backbones
Generating symmetric protein assemblies for nanotechnology or vaccine design
Designing high-affinity protein binders for therapeutic targets