[gmx-users] Re: Simulation Box break into 16 domains => Gromacs 3.3.3

[Justin A. Lemkul]

Chih-Ying Lin wrote:
>  
> Hi
>  From the Gromacs Manual,
> http://manual.gromacs.org/current/online/editconf.html
> editconf =>
> The box can be modified with options -box, -d and -angles. Both -box and 
> -d will center the system in the box.
>  
> so, without -c option, the protein is already centered in the box.
> and, there is no description in the manual that people cannot use -d and 
> -box simultaneously. 
>  

The manual cannot possibly warn against everything that people may try to do. 
Consider the function of these two options.  By using -d, editconf will 
determine a suitable box size based on the dimensions of the solute.  By using 
-box, you are specifying the box vectors, which completely negates the purpose 
of using -d.  Perhaps the language in the manual is not clear.  Both -d and -box 
may center the solute, but *not* by simultaneously using -box and -d, because 
they can counteract each other.  In the absence of a warning, something odd 
might be going on; we must eliminate this possibility in order to help you better.

I would still suggest using a properly-formed command, with or without -c, to 
make sure that nothing is breaking down there.  The workflow you posted before 
seems reasonable enough, so something early on is probably breaking down.  Is 
there any reason you minimized your protein without PBC?  What was the box size 
used in this EM step?  Usually an in vacuo EM is simply accomplished by using a 
huge box and plain cutoffs.  Solution EM would use a suitable box (with editconf 
-d *or* editconf -box) and better electrostatics methods.

-Justin

> I visualized the .gro file created by the editconf,  the protein is 
> centered in the box as I can see.
>  
> Thank you
> Lin
>  
>  
>  
>  
>  
> 
> Chih-Ying Lin wrote:
>  >
>  >
>  >
>  > Hi
>  > As I posted the command list earlier, to create the box
>  > editconf_mpi -f 6LYZ-EM-vacuum.gro -o 6LYZ-PBC.gro -bt cubic -d 0.75
>  > -box 6.0 6.0 6.0
>  >
>  >
>  > So, I think Justin's case is not the same as mine.
> Mark already pointed out that your command line is malformed.  You 
> cannot use -d
> and -box simultaneously.  They are mutually exclusive.  You are also not
> centering the protein in the box (you are not using the -c option).  So, in
> fact, what you're doing was exactly what my problem was - not centering and
> potentially defining the box incorrectly.
> I would suggest that you at least try rebuilding your system, because 
> the above
> command is certainly wrong.
> -Justin
>  > Thank you
>  > Lin

-- 
========================================

Justin A. Lemkul
Ph.D. Candidate
ICTAS Doctoral Scholar
MILES-IGERT Trainee
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin

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