[gmx-users] "comm-mode = Angular" gives error

[Justin Lemkul]

On 8/19/19 12:39 AM, Jorden Cabal wrote:
> Dear Users,
> I am trying to perform simulation of a very large and complex system and
> which requires me to make the simulation box rectangular cube instead (box
> dimension 24.5 nm * 12.5 nm * 12.3 nm) of perfect cuboid in order to
> minimize the number of atoms present in the system.
>
> I am facing the problem of interaction between periodic images due to
> rotational movement of the macro molecule. I have tried lowering the
> *nstcomm* (frequency of removal of center of mass velocity) to 10 and it
> seems it might work but the speed become very slow because it also requires
> *nstcalcenergy* to 10.
>
> When I use *comm-mode="Angular"* with *comm-grps="system" *it says "*too
> many warnings*" and the warning is
> *"Removing the rotation around the center of mass in a periodic system,
> this can lead to artifacts. Only use this on a single (cluster of)
> molecules. This cluster should not cross periodic boundaries*"
>
> *"Although this problem can be ignored using -maxwarn option in the grompp
> command. I want to know that if that is the correct thing to do?" *
>
> *Instead, *I have also tried to work this out by changing the* comm-grps=
> "Protein" *but then it throws two warnings first being the same as
> mentioned above and the second warning says "*Some atoms are not part of
> any center of mass motion removal group. This may lead to artifacts. In
> most cases one should use one group for the whole system*"
>
> Again this can be ignored using -maxwarn option but is that right to do?
> Please suggest me regarding this.

This is not the purpose of comm-mode/comm-grps, which account for net 
COM motion as a function of other errors in the simulation (Google the 
"flying ice cube effect"). You can't prevent your molecules from 
rotating in any physically realistic way. The simulation box needs to be 
larger to avoid these interactions, and comm-mode/comm-grps is not the 
way to get around it.

-Justin

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

Justin A. Lemkul, Ph.D.
Assistant Professor
Office: 301 Fralin Hall
Lab: 303 Engel Hall

Virginia Tech Department of Biochemistry
340 West Campus Dr.
Blacksburg, VA 24061

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

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