[gmx-users] comm removal in xy-direction only

[Diddo Diddens]

Dear all,

I have a question concerning the center-of-mass-motion removal in 
gromacs. Is it possible to remove the comm in x- and y- direction only, 
while still having periodic boundary conditions in all three dimensions?

Currently, I am simulating a polyelectrolyte complex in solution. The 
structure is a bit similar to that of a membrane, as the complex is 
periodic in x- and y-direction, whereas in z-direction, the solution 
lies above and below the complex. The comm is removed separately for the 
complex and the solution.

In a second step, I would like to apply an external force in 
z-direction. Since any comm removal in z-direction would undo the motion 
imposed by the external force, I was wondering if one could remove the 
comm in x- and y-direction only (preferably only for the comm-group 
"complex"). I noticed that this is automatically done for pbc=xy, 
however, I would like to know if this can also be achieved for pbc=xyz. 
If not, could someone please specify the exact location in the gromacs 
code, so I could implement this little feature myself?

Many thanks in advance.

Best regards,

Diddo Diddens