search

martinpacesa / BindCraft

User friendly and accurate binder design pipeline
MIT License
206 stars 38 forks source link

After hits, how to do the negative selection? #59

Closed finexzhi closed 3 days ago

finexzhi commented 3 days ago

I have two similar proteins A and B. The only difference in sequence is about 4 amino acids. I am trying to get a specific binder recognizing A but not B. I have used the different amino acids in A as hot spots. Finally, I got about 20 hits, which all bind to these "hot spot" amino acids. I want to know if these binders will have no or weak binding to B. What software or algorism I should use to evaluate their binding to B before I jump to the wet lab examination. Thanks.

martinpacesa commented 3 days ago

That will be very very hard for BindCraft to differentiate. We are currently developing a method for it, but this won't be public for a while.