[gmx-users] Re: Mdrun interrupted using a triclinc box

[Tsjerk Wassenaar]

Xie Yinghong,

First, forget the idea that trjconv has anything to do with the box in your
simulation. It is only concerned with the representation of your system
afterwards (for visualization). If your system crashes, it has nothing to do
with trjconv.

Second, forget the idea of a 'box' in your simulation. You have a single
simulation system, which is replicated over the three lattice vectors
defined by the nine numbers at the end of the .gro file. The shape of your
simulation system can be choosen freely (hexagonal prism, rectangular,
non-convex), the resulting infinite simulation system formed by the central
unit cell and its replicates will be the same.

Please DO read chapter three of the manual and have a look at Bekker, H. J
Comput Chem 18:1930 (1997) Also, check the Gromacs archives for anything to
do with PBC or 'system explodes'

If your simulation crashed, it has to do with what you feed to mdrun, and
what happens there. What you do afterwards doesn't make a difference. It may
be that you have a problem with the topology of your system. Also, it may be
that in setting up your system, you have chosen a set of lattice vectors,
which lead to overlap of atoms, which is always a good reason for a
simulation to crash.

In order to help you further, we would need to know if grompp gave warnings
during the generation of the .tpr file. Also, it would be helpful to see the
last few lines of the .log file from mdrun. You should check for yourself
whether your system had overlaps before starting the simulation, which you
can do by using genconf to replicate your system a few times.

btw. Reply to the Gromacs user list and not to individuals, as it will allow
more people to answer your questions and later read them in the archives.

Tsjerk

--

Tsjerk A. Wassenaar, M.Sc.
Groningen Biomolecular Sciences and Biotechnology Institute (GBB)
Dept. of Biophysical Chemistry
University of Groningen
Nijenborgh 4
9747AG Groningen, The Netherlands
+31 50 363 4336
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20051010/f4ce6796/attachment.html>