Hi,
I too had the same problem of reimaging in lipid system.
I think what Jason suggested works well in this case.
I tried in my system "center" command followed by "image origin center"
trajin xyz.mdcrd
trajout xyz_reimage.mdcrd
center :<entire lipid residue>
image origin center
Hope this helps
Shweta.
On Tue, Mar 11, 2014 at 5:15 PM, Jason Swails <jason.swails.gmail.com>
wrote:
> On Tue, 2014-03-11 at 06:37 +0000, Vijay Manickam Achari wrote:
> > Dear Dan, Jason, and David,
> >
> > Thanks for your prompt response. Well, I keep trying those commands
> > that you suggested but I don't get of what I want. I keep seeing the
> > layers of water like climbing a ladder. The same goes to those lipids
> > layers as well.
> >
> > As to understand what is going on during the processing of the
> > trajectories, I would like to investigate these commands one bye one
> > systematically.
> >
> > Start from first I tried to use the command
> >
> > ptraj topology.top << EOF
> > trajin abcdef01.traj
> > trajin abcdef012traj
> > ..
> > ..
> > ..
> > unwrap :1-1214
> >
> > trajout ABCD.nc netcdf
> > EOF
> >
> > I notice some of the water molecules become so big in size compare to
> > the other water molecules. I have attached a snapshot of that for your
> > view. It is really surprising me. Why few water molecules having some
> > length bonds?
>
> Are you using "orthographic" or "perspective" view in VMD? If the
> latter, then the "big" water molecules are just the ones that are closer
> to you. If you rotate the system, the "big" molecules will move
> backwards in the screen and appear smaller (but ones that have rotated
> to the front will now appear much larger).
>
> Also note that trajectories are stored in single precision in NetCDF
> format. Therefore, as the coordinates get larger (i.e., as waters
> diffuse farther and farther away), these numbers begin to lose precision
> and you may see strange artifacts as a result.
>
> In the end, the _only_ reason that I can think of that you would want to
> "unwrap" your trajectory would be in order to study diffusion properties
> (and even then you need to run in the NVE ensemble to get meaningful
> results, I believe).
>
> Good luck,
> Jason
>
> --
> Jason M. Swails
> BioMaPS,
> Rutgers University
> Postdoctoral Researcher
>
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Wed Jul 16 2014 - 04:30:02 PDT
