[gmx-users] Re: genbox nmol add water (Dallas Warren)

[Pim Schravendijk]

On Mon, 21 Mar 2005 gmx-users-request at gromacs.org wrote:
> 
> >how can I manage to put a certain number of water molecules in a box? I 
> >tried the genbox -nmol option with -ci spc216.gro, but it didn't work out 
> >due to "more than one residue". Can I convert the spc216.gro file so that 
> >it'll contain only one residue (and how) or is there a better way? I need 
> >to have the same number of molecules in a box, if I want to do free energy 
> >calculations, right?
> 
> copy the spc216.gro to something like spc.gro or h2o.gro, then text edit 
> the file, deleting all put the first three atoms, adjust the second line to 
> the correct number of atoms, and reduce the box size.  You can then use 
> genbox, but the code behind it isn't that good as you may end up with some 
> high energy overlaps. But with something as simple as water a quick energy 
> minimisation should sort that out.
> 
> Catch ya,
> 
> Dr. Dallas Warren

You could also choose not to use the -nmol option but instead choose the
boxsize size in such a way that you have a bit more water than wanted, and
remove the excess waters in the conf file (do not forget to adjust the
second line in the conf file here as well, and the SOL value at the end of
your topol.top). Make sure to equilibrate your box with pressure coupling
(when there is a lot of emtpy room use a very high pressure) long enough
to get the right density again before doing runs with this system. The
advantage of doing it this way is that you are using the box with 216
water molecules that are already more or less randomly oriented. If you
use only one water molecule to fill the box you will start with a
completely ordered box and equilibration will take a lot longer.

--
Pim Schravendijk - PhD Student
Max Planck Institute for Polymer Research
http://www.mpip-mainz.mpg.de/~schraven/