[gmx-users] octahedron box shape problem

THOMAS_STOCKNER stockn at kfunigraz.ac.at
Wed Oct 31 17:40:59 CET 2001


On Wed, 31 Oct 2001, Erik Lindahl wrote:

> thomas stockner wrote:
> 
> > Hi
> >
> > I am trying to generate an octahedron or dodecahedron box for a MD
> > simulation, but unfortunately I  am not succesful. I get only rectangular
> > boxes.
> >
> > I am using Gromacs version 3.0.5
> >
> > applying a commands like:
> > -editconf -f struc.g96 -o box.g96 -d 1.4 -bt octahedron -label A
> > -genbox -cp box.g96 -cs spc216.gro -o h2o.g96 -p struc.top
> > and creating a pdb file for visualizing with rasmol
> > -pdb2gmx -f h2o.g96 -o rasmol.pdb
> > the box I get is always recangular, independenty which options of editconf I
> > tryed (-bt; -angles; -box).
> > I don't understand, why the output of editconf regarding the box statistic
> > does not correspond to characteristics of the filled water box of the pdb
> > file visualised with rasmol.
> >
> > I cannot find the error I make. Could anyone help me to generate an
> > octahedron box.
> 
> Hi Thomas,
> 
> What does the last line of the .gro file look like? If there are nine numbers
> and more than three of them are non-zero the box is probably correct.


Hi Erik,

thanks for the hint. I already checked the possibility, that my viewer
could show me the box in a wrong way. Unfortunately is was not the case,
as I check with ngmx and i filled the box with water molecules for
visualizing its borders.


the last lines of the structure file are:
$ tail h2o.g96
 7098 SOL   OW     21890    6.587239742    5.620179653    4.878179550
 7098 SOL   HW1    21891    6.524239540    5.548179626    4.847179890
 7098 SOL   HW2    21892    6.680239677    5.584179878    4.878179550
 7099 SOL   OW     21893    0.252369881    6.039179802    4.415120125
 7099 SOL   HW1    21894    0.193369865    6.119179726    4.419119835
 7099 SOL   HW2    21895    0.267369270    6.004179955    4.508120060
END
BOX
    6.633870125    6.254472733    5.416532040
END

the output of editconf is:
.
.
    diameter    :  3.834               (nm)
    center      :  2.220  3.577  0.823 (nm)
    box vectors :  0.100  0.100  0.100 (nm)
    box angles  :  90.00  90.00  90.00 (degrees)
    box volume  :   0.00               (nm^3)
    shift       :  1.097 -0.450  1.885 (nm)
new center      :  3.317  3.127  2.708 (nm)
new box vectors :  6.634  6.634  6.634 (nm)
new box angles  :  70.53 109.47  70.53 (degrees)
new box volume  : 224.74               (nm^3)
.
> 
> 
> I don't know right off, but one possibility is that the gromacs box
> really is an octahedron but it doesn't show in your viewer program.
> 
> ngmx can show the different box types if you select it from one of
> the menus. Note however that the water molecules might not all
> be drawn within the box before you have performed any simulation,
> and that might cause it to look like the box isn't right.
> 
additionally i made a short md run in order to get a trajectory for using
ngmx. again the box is rectangular. when trying change the box shape to
octahedron in get the error message:
"Can not change rectangular box to truncated octahedron"

it seem like as a truncated octahedron box was never created. or could it
be that genbox changes again the border definitions??



cheers

thomas


-- 

  --
 ----/ /  _
 ---/ /  (_)__  __ ____  __
 --/ /__/ / _ \/ // /\ \/ /
 -/____/_/_//_/\_,_/ /_/\_\
                                    

-------------------------------------
Thomas Stockner
Institute of Chemistry
University Graz
Heinrichstr. 28
A-8010 Graz
Austria




More information about the gromacs.org_gmx-users mailing list