[gmx-users] Gromacs 2016 pull options

[Gmx QA]

Dear list

Could someone please clarify a few issues I have with the 2016 version of
the pull code. Or rather, explain why the below approach is not working.

I am trying to pull a small molecule though a bilayer. Initially, the drug
is "below" the membrane, i e has a smaller z-coordinate, so I want to pull
in the positive z direction through the membrane.

The following is a snippet of my mdp-file, with the relevant parts for the
pull specs.

pull                      = yes
pull-ngroups             = 2
pull-ncoords             = 1
pull-coord1-groups       = 1 2
pull-group1-name         = drug
pull-group2-name         = membrane
pull-coord1-type         = umbrella
pull-coord1-geometry     = direction-periodic
pull-coord1-vec          = 0 0 1
pull-coord1-rate         = 0.1

This I think should define two pull groups,one for the drug and one for the
membrane, and it should pull the drug molecule with pull-coord1-rate in
positive z. I am using direction-periodic since I expect to pull more than
half the box z length.

When I test this, running a 1 ns simulation, during which I expect the drug
to be pulled a significant way, nothing really happens with the com of the
drug, it just "sits" out in the water phase.

Please advice!
/PK