[gmx-users] This may be it, but what to do?

[Erik Lindahl]

Hi Marc,

The .gro format should contain:

1 line with title/comment
1 line specifying the number of atoms (n)
n lines with atom coordinates
1 line with the box specification

For a rectangular box, the box specification is just three free  
format floating-point numbers (x,y,z axes).

Gromacs can also work with triclinic boxes, which can be used to  
represent truncated octahedron or rhombic dodecahedron unit cells. In  
this case, the box is represented as a symmetric 3x3 matrix. To keep  
it compatible with the rectangular boxes, those box specifications  
contain six numbers, namely the elements

XX    YY     ZZ      XY      XZ      YZ


The most likely problem in your case is that you have specified too  
few/many atoms, and thus grompp isn't finding a reasonable box, but  
tries to generate one from the largest/smallest coordinate in each  
dimensions.

Good programs (VMD, Pymol) should be able show you triclinic boxes  
(we can export the triclinic box correctly to a pdb file), but some  
programs might just show you the diagonal components as a rectangular  
box.


Cheers,

Erik

On Aug 23, 2006, at 4:35 AM, Marc Charendoff wrote:

>
> Here's what I've been able to do so far . . .
>
>
> 1. Generated a  two molecule complex and saved as .pdb
> 2. Parsed the file out to retain relative coordinates
> of each molecule in separate .pdb files
> 3. Run both resulting molecules through PRODRG server
> to generate .gro files for GMX processing.
> 4. Merge molecules back together into a single complex
> .gro file (again, hoping the coordinates are
> preserved)
> 5. grompp resulting .gro for energy minimization run
> 6. run mdrun and watch my simulation explode
>
> When I visualize my complex.gro file with VMD,
> everything looks skewed (which now explains why my sim
> didn't work). Any idea what happened? When I view the
> PRODRG generated .pdb files, everything looks fine.
> Did PRODRG somehow skew my .gro files? Any guidance
> you could provide would be appreciated.
>
> Regards, Marc
> _______________________________________________
> gmx-users mailing list    gmx-users at gromacs.org
> http://www.gromacs.org/mailman/listinfo/gmx-users
> Please don't post (un)subscribe requests to the list. Use the
> www interface or send it to gmx-users-request at gromacs.org.
> Can't post? Read http://www.gromacs.org/mailing_lists/users.php