From: Axel Kohlmeyer (
Date: Fri May 28 2010 - 08:52:10 CDT

On Fri, 2010-05-28 at 12:40 +0200, Amir Hajiahmadi Farmahini wrote:
> Hi,

dear amir,

> I have simple questions regarding RDF and coordination number in VMD.
> Due to some problem I got confused by the way VMD calculates RDF and
> Coordination number.Unfortunately VMD user guide is not clear in this
> case .

there is only one way how to compute the atomistic g(r) and it is well
documented in text books, so why repeat it over and over again?

there seems to be a trend recently that people expect a software
documentation to be fully self-sufficient, i.e. also to contain
explanation of the underlying theory and the methods. it is important
to realize that the purpose of a software documentation is to explain
how to use it.

> 1- VMD asks for two selections to compute RDF. Which one is considered
> as central atom and which one is being counted around the central atom.
> I assumed that the second selection is counted around the first
> selection within r(max)distance. Is that correct?

mostly. VMD computes spherical pairwise atomistic RDF, so it looks
at the distribution of the atoms in selection 2 around the atoms in
selection 1. both selections can overlap or be identical. the
distribution is computed for each atom in selection 1 and then

> 2- In the Out put file third column (igr)is the coordination number and
> the fourth(isto) is the unnormalized histogram. Is that true?


> I guess the igr is the right number when we need to know the number of
> surrounding atom in vicinity of the atom of interest.

correct. you take the minima in the g(r) as reference points
and r_upper - r_lower for each region of interest gives you
the average number of atoms of selection 2 in that shell
around the central atom.

> Thanks in advance.
> Amir

Dr. Axel Kohlmeyer
Institute for Computational Molecular Science
Temple University, Philadelphia PA, USA.