[gmx-users] g_mindist, truncated octahedron and amber trajectory

[Berk Hess]

>From: cailliez <Fabien.Cailliez at ibpc.fr>
>Reply-To: Discussion list for GROMACS users <gmx-users at gromacs.org>
>To: gmx-users at gromacs.org
>Subject: [gmx-users] g_mindist, truncated octahedron and amber trajectory
>Date: Thu, 06 Jan 2005 16:50:02 +0100
>
>Dear all,
>
>I am very interested in using g_mindist tool from the GROMACS package to
>analyse
>the minimum distance between a solute and its image during a simulation.
>As I understand, g_mindist can read AMBER trajectory files when renamed
>with g87 extension.
>But I do not understand how g_mindist is working with box information.
>In AMBER trajectories, box information that is written at the end of
>coordinates only contains
>the length of the three vectors of the box (even when this is an
>truncated octahedron). The type of box
>is contained in the topology file.
>This does not seem to be written in the same way in g87 files, so when I
>use my AMBER trajectory,
>I think the images are created assuming that my box is orthorombic.
>I did not manage to solve this problem.

I assume AMBER, like GROMOS implements a truncated octahedron
using a cubic box with the corners cut off. Gromacs supports all possible
box shapes using triclinic unit cells. A single conformation can be 
converted
(not only changing the box line, but also rotating the whole configuration)
using editconf. The help (editconf -h or the online manual page) describes
how you can do this.

Berk.

_________________________________________________________________
MSN Search, for accurate results! http://search.msn.nl