[gmx-users] Protein moves out of box after several simulations, how to fix it?

[Justin A. Lemkul]

On 5/23/12 5:30 PM, Marc Hömberger wrote:
> Hallo,
>
> I have several (12) simulations running in which I simulate a protein
> in a water box (cubic 9x9x9). For all simulations I can  see the
> following:
> After a few simulations (total time of 7.6ns) the protein moves out of
> the water box - for all the twelve simulations. Since the following
> simulation is started with the output .gro file of the prior
> simulation the protein moves further and further out of the box.  I
> have no idea if this is a problem because of periodic boundaries or if
> this is a problem with my vmd installation. Can anyone give me a hint
> what I could to to fix this and not have to restart all the runs?
>

Your observations are consistent with 100% normal (and expected) behavior. 
There is no "outside" of a periodic box.  Check out the first two bullet points 
here and the explanation that follows:

http://www.gromacs.org/Documentation/Terminology/Periodic_Boundary_Conditions

> I am using a script I used before and I am doing everything like last
> time but last time I did not see this behavior of the protein. I am
> using vmd to look at the protein and its trajectory (commands see
> below)
>

Whether or not a molecule stays near the perceived "center" of the unit cell is 
simply a matter of probability.  Given sufficient simulation time, all proteins 
in water will cross at least one periodic boundary.  Such is the nature of 
diffusion.

-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

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