RF to refine PDB

[ZhuLvs]

I wish to utilize RF to refine a low-resolution single-chain PDB structure into a stable one. Which component should I employ?
thinks！

[roccomoretti]

If you already have a general topology you wish to maintain, then partial diffusion may be what you want to do.

If you're more interested in rebuilding certain areas from scratch, that's more motif scaffolding (but where the "motif" may be a substantial portion of the protein).

But this may depend on what you're attempting to achieve. RFdiffusion is really aimed at (de novo) backbone generation, so unless you can (re)frame your problem in terms of backbone generation, RFdiffusion may not be the tool to use. (That said, there are a number of tasks which can be reframed as backbone generation.)