search

JeffSHF / ColabDock

Code for ColabDock paper
Other
114 stars 12 forks source link

MvN restraints #17

Open PodevynLoris opened 10 months ago

PodevynLoris commented 10 months ago

Hello,

How should I obtain the .pkl file mentions on the config.py for MvN restraints ?

best, Lo

JeffSHF commented 10 months ago

Hi, there. The restraints should be obtained from wetlab experiments. If you just want to test the performance of ColabDock, you can extract restraints from native structures using the extract_rest.py script. (https://github.com/JeffSHF/ColabDock#restraints-sampling)

PodevynLoris commented 10 months ago

Thank you for your answer Jeff.

I am looking for the correct docking software to use. See I have an experimentally validated complex https://www.rcsb.org/structure/7F62 , my goal is to design a monomer (single chain structure) that would bind better to the receptor than the ligand found in the experimental complex (which is the CDR region of an antibody, so its made of two chains). SO I have generated some monomer that look like the antibody, but when I use a software to test the binding affinity (I use prodigy) of it in complex with the receptor , the binding affinities can be much BETTER (and this is due to unfair comparaison) like sometime my monomers have clashes with the receptor and this increase the score.

So I was wondering : By inputing to ColabDock my complexes (and the template being the experimental one) and by restraining correctly , could I avoid these kind of clashes and have more relevant binding affinities ?

Best, and thank you for your work.

LO

JeffSHF commented 10 months ago

I think ColabDock can solve the clashes, if reasonable restraints given. You can also try other relaxation protocols, such as rosetta.