Ability to redesign existing scaffolds to create binding interfaces in comparison to de novo binder design

[alllirik]

Hello!

Thank you for publishing your work! I'm currently experimenting with your software, and I have some ideas in mind, but I am not sure to what extent they are practical and possible with RFdiffusion. As I am new to the protein design field, feel free to correct any assumptions I may have made; I would greatly appreciate any guidance.

I have a target protein and a highly stable scaffold protein; however, the scaffold protein was not designed to bind to this specific target, so no actual binding motif is present. Is it possible (and effective) to use the RFdiffusion protocol, such as motif scaffolding, to redesign some parts of the existing scaffold that face the target hotspot to create a binding interface? Or would it be more effective to use partial diffusion/fold conditioned binder design to create new, but structurally similar scaffolds? If I understand correctly, the latter approach will cause loss of the binder sequence, which might lead to a possible loss of stability compared to the original scaffold.
I apologize if I am missing something about presented pipelines.

Thank you!

[roccomoretti]

RFDiffusion is more about designing the backbone topology/structure. (It can design a backbone structure in the context of a pre-specified motif, though.) It should be able to design a backbone topology which is compatible with binding to a particular target, but is probably not all that useful if you already have a particular scaffold structure you want to use. -- It might be useful if you wanted to in-paint loops or otherwise wanted to reorganize the backbone near the binding interface while keeping the rest fixed, as there are ways to pre-specify parts of (or almost all of) the context in which you're doing the redesign.

If you're looking for designing the sidechain identities of a fixed backbone (including redesigning the backbones output by RFDiffusion), you probably want to look into ProteinMPNN, which can be set up to redesign the sequence of either some or all of the protein.

But you're probably correct that redesigning the backbone will "invalidate" the current sequence, and may require a new sequence to support the new structure. ProteinMPNN is pretty good at figuring out that new sequence, though like most protein design processes, the success rate is a fair bit lower than 100%.

[alllirik]

Thank you for your detailed answer! I will look into described methods to redesign the scaffold, maybe something will work out.