[gmx-users] periodic-molecules = yes and movie

[Justin Lemkul]

On 8/7/18 8:51 AM, Alex wrote:
> Thanks.
>
> On Tue, Aug 7, 2018 at 8:30 AM Justin Lemkul <jalemkul at vt.edu> wrote:
>
>>
>> On 8/7/18 8:23 AM, Alex wrote:
>>> Dear all,
>>> In the simulation of a system in which there is slab, I did not use
>>> "periodic-molecules
>>> = yes" unfortunately, hence I can not make a good movie out of the
>> This is not some magic setting that will make nice movies. Note that a
>> "periodic molecule" is different from just normal PBC. Periodic
>> molecules are "infinite" molecules like sheets, nanotubes, etc. that
>> have bonds between central image atoms and periodic copies of atoms. If
>> you needed such a setting and did not use it, likely your simulation
>> would have collapsed.
> No, the simulation did not blow up and worked nicely. As far as I
> understood and in general the periodic-molecule = yes should be used when a
> surface is connected to it's image in the neighbouring cell, mine is silica
> slab, and if the the slab is defined only by non-bonded interaction like
> most of the metallic slab the periodic molecule = NO should work finely in
> spite of being infinite.

Correct. "Periodic molecules" only matter for bonded interactions.

>
>>> trajectory using VMD, the slab move during the movie. The simulation is
>>> around 300 ns and I do not want to just waste the CPU time, so, I wonder
>> if
>>> there is still any solution for that to make a good movie out of that
>>> trajectory?
>> Many of the normal trjconv routines could likely help here. Accounting
>> for PBC effects like these is a standard first step in post-processing
>> any trajectory.
>> Actually what I see is the result of two below comments applied over the
>> initial trajectory:
>>
> gmx_mpi trjconv -f prd.xtc -s eql1.gro -o nojump-skip-30-prd.xtc -center
> -pbc nojump -n input/index.ndx -tu ns -skip 30   < choice : salb, system
> gmx_mpi trjconv -f nojump-skip-30-prd.xtc -s prd.tpr -o
> nopbc-nojump-skip-30-prd.xtc -center -pbc whole -ur compact -n
> input/index.ndx -tu ns < choice : salb, system

If the surface appears to be shifting laterally, then a translational 
fit is probably what you want.

-Justin

-- 
==================================================

Justin A. Lemkul, Ph.D.
Assistant Professor
Virginia Tech Department of Biochemistry

303 Engel Hall
340 West Campus Dr.
Blacksburg, VA 24061

jalemkul at vt.edu | (540) 231-3129
http://www.thelemkullab.com

==================================================