[gmx-users] Ion positioning 2

[Lieven Buts]

Anton Feenstra wrote:
> Is this perhaps only an apparent problem, meaning that due to the
> periodic boundaries, the ions are actually near the protein, but
> only appear to be 'far away' because they happen to end up at the
> other end of the box? To assess this, you should keep in mind that
> the water box using '-bt dode' is also brick-shaped, but that the
> stacking of the brick determines that actual box shape, i.e. the
> bricks are stacked with a shift rather than directly on top of each
> other.

I was starting to wonder about the shape of the water box :-).
I still don't completely understand this "stacking with shifts"
but I'll look into it.

> # trjconv -f conf.gro -s conf.gro -ur compact -pbc inbox -o compact.gro

Cool. This looks better (<http://ultr.vub.ac.be/~lieven/ions3.png>).
Some of the ions are still far from the protein, but they are no longer
in an "unnatural" geometric arrangement as they seemed to be in
the brick-shaped view (<http://ultr.vub.ac.be/~lieven/ions.png>).

-- 
Lieven Buts
Vrije Universiteit Brussel