[gmx-users] center peptide in simulation box

[Mark Abraham]

On 9/03/2010 10:13 PM, politr at fh.huji.ac.il wrote:
> Dear Gromacs users and developers,
> I'm trying to visualize .trr from md simulation using VMD.
>
> I see that starting from some point (probable as a result of a peptide
> drift ) my peptide splits and it seems like one part appears at one side
> of the box and the other on the opposite side. Both parts are connected
> with a very long bond. I want to center the peptide in the box and of
> course I want all the molecules to be wrapped accordingly. I checked the
> mailing list and I found that I can use trjconv (trjconv -f *.xtc -s
> *.tpr -o center.xtc -center -pbc nojump). I tried to use this command
> with .trr (trjconv -f *.trr -o center.trr -center -pbc) as a result the
> peptide is still not centered and all the water molecules diffuse out of
> my rectangular box. Am I doing something wrong? I'm trying to center the
> molecule not only for visualization but also because I want to analyze
> the simulation using VMD and scripts I made for analysis of NAMD
> simulations (I worked with NAMD for a long time and I need to work with
> gromacs because of SPCE water).

OK, well keep playing with the trjconv options and choosing sensible 
groups for centering until you find something that pleases you. You may 
need to do two different trjconv operations - one to center the peptide 
and one to group the waters in the same periodic cell as the peptide.

> Is it possible to perform analysis with
> VMD (I need commands like measure distance, measure angle, cut molecules
> at some radius from predefined residue and pbcwrap around some molecule
> or peptide)?

Yup.

Mark