From: Krishnan, Marimuthu (krishnanm_at_ornl.gov)
Date: Thu Apr 23 2009 - 12:11:07 CDT

Axel,

"please don't mix up vdw interactions and vdw radii. bonded atoms still
have vdw radii, it is just that the vdw interaction is excluded, to
simplify the parametrization of bonded interactions."

What I meant to say is that van der Waals spheres of bonded atoms overlap (for example,
carbon-carbon bond length is 1.53 Angstroms while van der Waals radius of carbon is
around 1.7 Angstroms (http://en.wikipedia.org/wiki/Van_der_Waals_radius)).
It is my understanding that use of vdW radii for bonded pair of atoms can lead to
an overestimation of total volume occupied by that atoms of interest
(the chemial group+near neighbors). This will result in some errors in the local
packing density.

Krishnan
 

-----Original Message-----
From: Axel Kohlmeyer [mailto:akohlmey_at_cmm.chem.upenn.edu]
Sent: Thu 4/23/2009 12:27 PM
To: Krishnan, Marimuthu
Cc: vmd-l_at_ks.uiuc.edu
Subject: Re: vmd-l: local packing density around a chemical group (say, a methyl group or a side chain)
 
On Thu, 2009-04-23 at 11:15 -0400, Krishnan, Marimuthu wrote:
> Dear friends,
>
> I would like to calculate the local packing density around a methyl group (or a side chain) of a protein.
> The MD trajectory, which was generated using NAMD, of this protein is available. I would appreciate if
> someone could direct me to some existing codes/scripts that can do this task.
>
> I would be happy to write a simple code to do this job but I need some clarifications on this topic.
> To compute local packing density, I need to know the vdW radii of the chemical group and its neighbors.
> I am wondering about contributions from bonded atoms (There is no van der Waals interactions between
> atoms involved in a covalent bond, angle and dihedral). The questions are:
> (1) How do I get vdW radii of the chemical group( or atoms) and its neighbors?

VMD has deliberately written without any chemical knowledge and
guesswork. it will process whatever information you feed it and
otherwise will use more-or-less useful defaults. you can access
atom data via functions generated with the atomselect command.
those can set and retrieve per atom properties, including radius,
mass, charge, name, type, element and so on.

> (2) How do I take care of bonded pair of atoms in the computation of local packing density?

please don't mix up vdw interactions and vdw radii. bonded atoms still
have vdw radii, it is just that the vdw interaction is excluded, to
simplify the parametrization of bonded interactions.

cheers,
   axel.

>
> I would appreciate your help and suggestions.
>
> Thanks,
> Krishnan
>
> *************************************************
>
> Dr. Marimuthu Krishnan
> Postdoctoral Research Scholar,
> Center for Molecular Biophysics,
> Oak Ridge National Laboratory,
> Building 6011,
> PO Box 2008 MS6309,
> Oak Ridge, TN 37831-6309,
> Tel : 001-865-574-6123
> Fax : 001-865-576-7651
> E-mail : krishnanm_at_ornl.gov
> For more details : http://cmb.ornl.gov/group/kgh
> *************************************************
>

-- 
=======================================================================
Axel Kohlmeyer   akohlmey_at_cmm.chem.upenn.edu   http://www.cmm.upenn.edu
   Center for Molecular Modeling   --   University of Pennsylvania
Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323
tel: 1-215-898-1582,  fax: 1-215-573-6233,  office-tel: 1-215-898-5425
=======================================================================
If you make something idiot-proof, the universe creates a better idiot.