[gmx-users] shift in trajectory

[Justin Lemkul]

On 10/17/19 8:04 AM, Amin Rouy wrote:
> Dear Justin,
>
> I have a two-component system (they demix), which are grouped as  'A' and
> 'B' from beginning of the simulation. Now my trajectory is asymmetric with
> respect to the center. I can make it symmetric with gmx trjconv flag -shift
> manually, but I need to apply the flag -center to get the 'A' component
> sitting exactly at the center of the box.
>
> So having my index file, what I do is:
>
> 1- *gmx trjconv -f NPT.xtc -o NPT-centered.xtc -pbc none -n index.ndx*
>
> and I select whole system.

Making molecules whole should always be the first step. I disagree with 
the previous advice you were given to use this PBC option first.

>
> 2- *gmx trjconv -f NPT-centered.xtc -o NPT-centered2.xtc -n index.ndx
> -center*
>
> and I select group 'A' for Select group for centering, and whole system for
> Select group for output.
>
> 3- *gmx trjconv -f NPT-centered2.xtc -o NPT-centered3.xtc -n index.ndx -pbc
> mol -s NPT.tpr *
>
> and here I select whole system.
>
>
> What I get is the same trajectory as I had in the beginning, no change at
> all. I can see it with vitalizing and moreover from gmx density outcome.

Do you get a clear phase separation with a layer of A and a layer of B 
in the simulation? In steps prior to separation, it will not be apparent 
that the molecules are actually centered, but you can obtain their COM 
coordinates with gmx traj -ox -com and confirm that they correspond to 
the box center.

-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

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