[gmx-users] Umbrella sampling PMF, drug and membrane with pull_geometry=cylinder

[J B]

Hi.
I am trying to the the PMF of transferring a drug molecule from the aqueous phase to a lipid bilayer.
In order to minimize the effects of membrane undulations I tried to use pull_geometry=cylinder.
Whenever I run a short simulation with umbrella sampling starting from a configuration with the drug molecule way above the bilayerthe drug molecule is pulled into the bilayer rather rapidly. It appears as if the distance between the cylinder's COM and the drug molecule's COM is supposed to be zero since it is pulled to the center of the membrane. 
Since I apply an umbrella potential I find this weird and I would of course like the molecule to"stay put" at that current z-position. Probably I set up the pull code wrong. Here are my options:
pull                    = umbrellapull_geometry     = cylinderpull_dim             = N N Ypull_ngroups    	 = 1pull_rate1      	 = 0.0pull_k1         	 = 1000pull_start            = yespull_init1             = 0.0pull_vec1            = 0.0 0.0 1.0pull_pbcatom0     = 0pull_ngroups        = 1pull_group0         = DPPCpull_group1          = drug
Have not decided the radii of the cylinder yet, but that is besides the point here.
I run with the 4.5.5 version and cannot figure what the problem is. 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20120428/435a2c4f/attachment.html>