Using steered MD to select near-native from decoys of docking models – User discussions

GROMACS version: 2022.3
GROMACS modification: Yes/No

Hello, I wish to use steered MD to select near-native antibody-antigen docking pose from a bunch of decoy docking poses generated using ZDock.
My assumption is that the near-native structure would bind stronger than the decoys, and either the energy / force used to dissociate the two proteins can be utilized to indicate the difference.
I followed Justin Lemkulā€™s tutorial for umbrella sampling.

I can find the force from the pullf.xvg, but is there a way to calculate the energy put in to the pulling simulation without running umbrella sampling simulation for each window and running wham?
Does the area under curve of the pull force curve indicate anything?

Thanks in advance.

Read more here: Source link