[gmx-users] Cylinder option in pull code

[O'Neill, David]

Hi all,


I am using the pull code in Gromacs 4.5.4, to constrain the distance (z) between a small molecule and a lipid bilayer. I have included the pull_geometry = cylinder option as I wanted to limit the small molecule's movement in the x and y directions.  My understanding was that the radius of the cylinder would provide such a limitation, but after checking the trajectory and plotting  the coordinates of the COM over time (using g_traj), it is clear that the molecule is constrained in z but still has complete freedom in x and y.


Is there something wrong in my setup? Or am I misunderstanding how this cylinder option works and what it is designed for?


Pull parameters are listed below.


; Pull code
pull            = constraint
pull_geometry   = cylinder
pull_dim        = N N Y
pull_r1         = 1
pull_r0         = 1.5
pull_start      = yes
pull_ngroups    = 1
pull_group0     = A
pull_group1     = FRE2
pull_init1      = 0
pull_vec1       = 0 0 1
pull_rate1      = 0
pull_k1         = 1000      ; kJ mol^-1 nm^-2
pull_nstxout    = 1000      ; every 2 ps
pull_nstfout    = 100      ; every 0.1 ps

Any help or advice would be much appreciated.

Thanks,

David