[gmx-users] Visualization problem of bromo atom connected to an aromatic ring due to gromacs or VMD?

[Justin A. Lemkul]

Meyer-Almes, Franz-Josef, Prof. Dr. wrote:
> Dear all,
> 
>  
> 
> I have two question:
> 
>  
> 
> 1) Are the atom types I chose for the use of OPLS force field correct?
> 
>  
> 
> I use gromacs 4.0.7 with opls force field under Ubuntu Linux to simulate 
> enzyme drug interaction. Since the Dundee Prodrg Server converts PDB 
> files of drug structures just in an .itp file compatible with gromos 
> forcefields, I had to change the atom types as follows:
> 
>  

Simply changing the atom types is not likely an appropriate solution.  OPLS-AA 
is an all-atom force field, in which all H should be represented.  You have a 
grand total of 3 H atoms in this structure.  After a quick look at your 
structure, this is definitely not correct.

Also note that the charges that PRODRG generates fall somewhere between 
"marginal" and "totally incorrect" for even the GROMOS force fields.

<snip>

> 2) Visualization problem of VMD?
> 
> The second question is related to the first one:
> 
> I use VMD 1.8.7 under Windows XP. When I try to visualize above 
> mentioned Drg.gro file using VMD, the bromo atom appears as a tiny point 
> and is not connected to C2 as it should. When I convert the Drg.gro file 
> to a pdf-file by the command editconf, the pdf-structure displays the 
> bromo atom correctly connected to the aromatic ring. In addition, during 
> the course of an mdrun over 1 ns the relative position and the distance 
> of the bromo atom to C2 remained unchanged. The distance between BR1 and 
> C2 in the final mdrun after some relaxation was constantly 1.89 Angstrom 
> which is realistic. Is my observation a visualization problem of VMD or 
> a serious problem with gromacs which I do not see.
> 

I don't see how the problem is with Gromacs - you say that the output of mdrun 
is reasonable.  If you've defined a bond between two atoms, the bond will be 
there.  The topology is correct, VMD's interpretation of the molecule has no 
such guarantee.

> If it is a problem with VMD, Is there a remedy for the incorrect 
> visualization of the drug molecule?
> 

VMD guesses where bonds should be based on inter-atomic distances.  There may be 
ways around this effect.  Many have been mentioned on this list, so a few 
minutes of poking around in the archive should turn up some useful information. 
  There may also be some information in the VMD mailing list archive or in the 
VMD manual.

-Justin

>  
> 
>  
> 
> Thank you in advance,
> 
>  
> 
>  
> 
> Franz
> 
>  
> 
>  
> 
>  
> 
> F.-J. Meyer-Almes
> 
> Hochschule Darmstadt
> 
> Fachbereich Chemie- und Biotechnologie
> 
> Schnittspahnstr. 12
> 
> 64287 Darmstadt
> 
> Email: franz-josef.meyer-almes at h-da.de
> 
> Tel.: ++49-(0)6151-168406
> 
> http://www2.h-da.de/cub/index.php?id=177
> 
>  
> 

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

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

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