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

Jochen Hub jhub at gwdg.de
Mon Apr 30 14:12:45 CEST 2012


we had similar trouble some time ago, and we could solve it by using a 
good PBC atom. Make sure that your pbc_atom is in the very center of 
your membrane.

Btw, using "distance" to compare with "cylinder" is a bad test, since it 
does something very different (the distance is the same at z=com-1 and 
z=com+1. I would use "direction" or "position" for your test. You can 
also use position/direction for the umbrella simulations, you'll 
probably get similar results to the geometry "cylinder", but you'll also 
have to make sure your pbc_atom is fine.


Am 4/29/12 9:20 PM, schrieb J B:
>  > Does the cylinder geometry work if you use "pull_start = no" and
> pull_init1
>  > equal to your starting (restraint) distance? I'm just trying to go
> through all
>  > of the iterations of what might be failing. In theory, what you're
> doing is
>  > fine but I'd be interested to know if there was something failing in
> the input
>  > parsing such that pull_start is not being properly applied. The
> screen output
>  > of grompp will also tell you what it believes the restraint distance
> is. If it
>  > prints 0, then there's a problem with input processing.
>  >
>  > If the problem is reproducible in the latest Gromacs version (4.5.5),
> please
>  > file a bug report on redmine.gromacs.org so that it can be
> investigated. Please
>  > provide .tpr file of the system so that the developers can run what
> you are
>  > trying to do and see.
>  >
>  > -Justin
> It works if I set pull_start to the initial distance given by g_dist
> things work just fine.
> So this will be the way for now at least.
> Guess that there could be an error in the inout parsing then.
> Thanks a lot for your help, really appreciate it.

Dr. Jochen Hub
Computational Molecular Biophysics Group
Institute for Microbiology and Genetics
Georg-August-University of Göttingen
Justus-von-Liebig-Weg 11, 37077 Göttingen, Germany.
Phone: +49-551-39-14189

More information about the gromacs.org_gmx-users mailing list