[gmx-users] Position restraints umbrella sampling

[Emil Marklund]

Hi,

I am trying to calculate PMFs for a number of protein-protein complexes, doing pulling and Umbrella sampling mainly following the protocol found in Justins tutorial.

In my case it feels like I should remove position restraints in the umbrella sampling on the peptide that is kept stationary with pos. restr. in the pulling. And that also seems to be the consensus when reading older messages in this mailing list, that positon restraints should not be used in the umbrella sampling if you have no special reason to keep them (that the umbrella potential is enough to maintain COM separation). I am however a little confused since I can also find examples in the litterature where restraints seems to have been kept during umbrella sampling for reasons that is not always clear to me...

Should I in this case remove positons restraints in the umbrella sampling? What does setting pull_coord1_type = umbrella actually do, is the COM distance between the proteins restrained or is the COM of each protein restrained to a position in space? If it is the first scenario can't you end up with problems if the proteins start to rotate around there joint center of mass and thereby aproaches the edge of the box?

Best,
Emil Marklund
PhD Student in Molecular Biotechnology
Elf lab
Department of Cell and Molecular Biology
Uppsala University