[gmx-users] still problems with dodecahedron box and editconf

[Sebastian Kruggel]

Dear gmx-er,

 

I have problems creating (or visualizing) a dodecahedron box. I found
simular

problems in the mailing list, where running mdrun is supposed to help - but
this does't work in my case.

I generated my box with the following commands

 

> pdb2gmx -f $PROT.pdb -o $PROT.gro -p $PROT.top -ff oplsaa -water spce

> editconf -bt dodecahedron -f $PROT.gro -o $PROT.ec.pdb -d 0.9

> genbox -cp $PROT.ec.pdb -cs -o $PROT.b4ion.gro -p $PROT.top

 

> grompp -f $MDP/em.mdp -c $PROT.b4ion.gro -p $PROT.top -o $PROT.b4ion.tpr

> genion -s $PROT.b4ion.tpr -o $PROT.b4em.gro -nname Cl -nn 1 -g genion.log

 

# and after editing $PROT.top

 

> grompp -f $MDP/em.mdp -c $PROT.b4em.gro -p $PROT.top -o em.tpr

> mdrun -v -s em.tpr -o em.trr -c $PROT.b4pr.gro -e em.edr -g em.log

 

if I look at the tpr/trr with ngmx or chimera, the box is cubic and a very
big

part of the protein is out of the box - I found in the mailing lists that
this

shouldn't be a problem because of the periodic boundary conditions (even if

the manula says: 'Both -box and -d will center the system in the box'). 

 

I had a look at the tpr-file to see if it's maybe just a problem of

visualisation. I used gmxdump to look at the topology and found:

 

> gmxdump -s pr.tpr | grep box

 

Reading file pr.tpr, VERSION 3.2.1 (single precision)

   box (3x3):

      box[    0]={ 8.77170e+00,  0.00000e+00,  0.00000e+00}

      box[    1]={ 0.00000e+00,  8.77170e+00,  0.00000e+00}

      box[    2]={ 4.38585e+00,  4.38585e+00,  6.20253e+00}

   boxv (3x3):

      boxv[    0]={ 0.00000e+00,  0.00000e+00,  0.00000e+00}

      boxv[    1]={ 0.00000e+00,  0.00000e+00,  0.00000e+00}

      boxv[    2]={ 0.00000e+00,  0.00000e+00,  0.00000e+00}

 

So the box really looks dodecahedron in the first part, but what about the 0
0

0 boxv ?!?

 

Maybe somebody can help me figuring out if my box is 'really' dodecahedron?

Thanks in advance!

 

Sebastian

-------------- next part --------------
Dear gmx-er,
 
I have problems creating (or visualizing) a dodecahedron box. I found simular
problems in the mailing list, where running mdrun is supposed to help - but this does't work in my case.
I generated my box with the following commands
 
> pdb2gmx -f $PROT.pdb -o $PROT.gro -p $PROT.top -ff oplsaa -water spce
> editconf -bt dodecahedron -f $PROT.gro -o $PROT.ec.pdb -d 0.9
> genbox -cp $PROT.ec.pdb -cs -o $PROT.b4ion.gro -p $PROT.top
 
> grompp -f $MDP/em.mdp -c $PROT.b4ion.gro -p $PROT.top -o $PROT.b4ion.tpr
> genion -s $PROT.b4ion.tpr -o $PROT.b4em.gro -nname Cl -nn 1 -g genion.log
 
# and after editing $PROT.top
 
> grompp -f $MDP/em.mdp -c $PROT.b4em.gro -p $PROT.top -o em.tpr
> mdrun -v -s em.tpr -o em.trr -c $PROT.b4pr.gro -e em.edr -g em.log
 
if I look at the tpr/trr with ngmx or chimera, the box is cubic and a very big
part of the protein is out of the box - I found in the mailing lists that this
shouldn't be a problem because of the periodic boundary conditions (even if
the manula says: 'Both -box and -d will center the system in the box').
 
I had a look at the tpr-file to see if it's maybe just a problem of
visualisation. I used gmxdump to look at the topology and found:
 
> gmxdump -s pr.tpr | grep box
 
Reading
file pr.tpr, VERSION 3.2.1 (single precision)
  
box (3x3):
      box[    0]=
{ 8.77170e+00,  0.00000e+00,  0.00000e+00}
      box[    1]=
{ 0.00000e+00,  8.77170e+00,  0.00000e+00}
      box[    2]=
{ 4.38585e+00,  4.38585e+00,  6.20253e+00}
   boxv (3x3):
      boxv[    0]=
{ 0.00000e+00,  0.00000e+00,  0.00000e+00}
      boxv[    1]=
{ 0.00000e+00,  0.00000e+00,  0.00000e+00}
     
boxv[    2]=
{ 0.00000e+00,  0.00000e+00,  0.00000e+00}
 
So the box really looks dodecahedron in the first part, but what about the 0 0
0 boxv ?!?
 
Maybe somebody can help me figuring out if my box is 'really' dodecahedron?
Thanks in advance!
 
Sebastian