[gmx-users] Large VCM error for a POPC bilayer simulation

[himanshu khandelia]

Hi Folks,
I was trying to make a large POPC system from Tieleman's 128 lipid assembly.
Everything went well until I tried to run dynamics. While trying to run
dynamics, I got a large velocity for the center of mass (Large VCM) error.
The log file, mdp file, and the top file are attached as a suse .zip

Here is what I did:

- replicated the system thrice to make is 512 lipids (this was done in VMD),
and translated replicas along the bilayer plane accordingly

- to remove bad contacts, minimized it in GROMACS (steep minimization)

- to remove bad contacts farther, minimized using CG
 (The system looks fairly reasonable after minimization on visual
inspection)

- tried to assign velocities at 50 K and run NPT dynamics, with water
molecules constrained along the bilayer normal (z)

In the last step, the system crashes after a few hundred steps, reporting a
large VCM. A look at the pdb files reveals that the system is exploding with
large vacuum regions appearing.

THE BOX CENTER WAS VERY FAR FROM THE GEOMETRIC ORIGIN. The box is  ~ 13 x 13
x 6.7, and the center is at 6.3, 6.3, 3.5. So I also tried the following
before running dynamics:

- used editconf -c to center the box, before running dynamics. But this only
moved the center by less than 5 angstroms. Why ?
- used editconf -c -center 0 0 0 as a variant of the above.

The above did not solve the problem.
Please let me know if you want me to send you more details ?

Thanks for the help,

-Himanshu
MEMPHYS,
SDU, Denmark
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20071017/258660b3/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: files.zip
Type: application/zip
Size: 4858 bytes
Desc: not available
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20071017/258660b3/attachment.zip>