Re: [AMBER] Error while running tleap for the ligand containing two chlorine atoms

From: Jason Swails <jason.swails.gmail.com>
Date: Fri, 20 Feb 2015 08:08:12 -0500

On Fri, Feb 20, 2015 at 7:33 AM, Shreeramesh <shreeramesh.gmail.com> wrote:

> Dears Sir,
>
> I attached the "lig.mol2: file which contains two chlorine atom. While
> running tleap "lig.mol2" was combined with another ligand "lig2.mol2".
>
> After the preparation of complex (lig. mol2 and lig2.mol2), the chlorine
> atoms are converted into carbon and the bonds between chlorine and carbon
> are broken
>

​tleap does not delete bonds. The bonds are clearly defined in the mol2
file as bonds 10 and 11 (between atoms 7-8 and 7-9). As a result, the
bonds WILL be in your final topology file. You are visualizing a PDB file
in a visualization program that draws bonds based on simple distance
criteria, as Hannes pointed out. As a result, these bonds do not
necessarily reflect the bonds that are defined in your prmtop file.

If you want to make sure that the bonds are in your final structure, you
need to visualize with your *prmtop* file, not your PDB file. VMD, for
instance, only shows bonds when the distance is below a certain value for
PDB files, but shows all bonds defined in the prmtop file when you use that.

And why do you say that the chlorine was converted to Carbon?

I did not find the chlorine atoms in the text file of complex:
> "com_solvated.pdb"
>

​They are exactly where they are in the mol2 file: atoms 8 and 9.

ATOM 7 C1 CLM 1 -0.680 2.617 1.647 1.00 0.00
ATOM 8 Cl1 CLM 1 -1.158 3.504 0.160 1.00 0.00
ATOM 9 Cl2 CLM 1 -2.089 2.280 2.703 1.00 0.00



> I checked with different tools for visualization, the problems remains the
> same
>

​Again, use the prmtop file to determine whether bonds exist or not. You
can also use ParmEd to check that bonds exist:

parmed.py -p com_solvated.top

> printBonds .Cl=

​HTH,
Jason

-- 
Jason M. Swails
BioMaPS,
Rutgers University
Postdoctoral Researcher
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Feb 20 2015 - 05:30:03 PST
Custom Search