From: Axel Kohlmeyer (
Date: Wed Jun 10 2020 - 23:59:10 CDT

On Thu, Jun 11, 2020 at 12:26 AM Rameswar Bhattacharjee <> wrote:

> Hi all,
> I have a simulated periodic box of pentanol with very few amounts of
> water. I have plotted the g(r) for OW-OW for the system using vmd and I got
> the first peak at a very high value of g(r) (~80). It looks very odd to me.
> Is there a normalization issue (bulk density) as I have few water
> molecules?

whether you get correct normalization depends on whether you feed the g(r)
plugin correct data, e.g. with the correct boundary box.

if you have very few water molecules, then the peaks *must* have larger
absolute values than for a (pure) bulk system.
please think about what it is that you are computing!
g(r) is the ratio of the probability of finding an OW atom at distance r in
your system relative to that of an ideal gas.
the fewer OW atoms you have the smaller the normalization factor is that
you have to divide by. now what happens, if you divide the same number by a
large number versus a small number?

> I am using lammps for the simulation. Any suggestion will be highly
> appreciated. Thank you.

there is not enough information to give an authoritative answer. one can
only speculate.
however, the measure gofr command in VMD has been thoroughly tested over
many years (and several bugs found and fixed (some of them quite subtle)),
it gives consistent results with other well testing implementations and is
used by many people. that makes it now very unlikely that there are any
significant bugs remaining. it is much more likely that either your
interpretation of the data is not correct or your computation was not
correctly done.


> --
> *Regards*
> *Rameswar*
> *********************************************************************
> *Dr. Rameswar Bhattacharjee*
> University of Delaware
> *Newark, DE-19716*
> *********************************************************************

Dr. Axel Kohlmeyer
College of Science & Technology, Temple University, Philadelphia PA, USA
International Centre for Theoretical Physics, Trieste. Italy.