[gmx-users] Reducing trr storage volume

[Justin Lemkul]

On 1/4/13 11:09 AM, Shima Arasteh wrote:
> Dear users,
>
> I'm simulating a system of protein/water/popc/ions. The peptide is inserted through the transmembrane, and the peptide acts as a channel.
> In order to reduce the trajectory storage volume, I need to use xtc-grps in .mdp file. In my system, I'd rather to store the positions of peptide,ions and waters "near" the peptide (I mean a specific region near the peptide: entrances of peptide + through the peptide).The large amount of water molecules positions exist up and down the bilayer, are not expected to be saved.
> I'm wondering if it is possible to chose a specific region in xtc-groups? Or any other solution to reduce the storage volume of trr file?
>

I doubt you can achieve such a group consistently.  Water molecules will diffuse 
over time.  Index groups provided to grompp can only be specified to be a fixed 
size and cannot be dynamically created to satisfy any geometric criterion during 
mdrun.  Thus, your initial group may indeed represent the desired molecules, but 
over time, the waters within the group will (likely) drift away and may not be 
anywhere near the original position.

If your system contains a lot of water that you deem unnecessary, perhaps you 
should consider the manner in which the system was constructed...

The only way to decrease the size of a .trr file is to decrease nstxout, 
nstvout, and/or nstfout.  For various technical reasons, there is no way to save 
only a subset of atoms in .trr format.  Refer to previous discussions on this topic.

-Justin

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

Justin A. Lemkul, Ph.D.
Research Scientist
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin

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