[gmx-users] Radial distribution function (RDF) in cylindrical coordinates

[Andrew DeYoung]

Hi,

I have a system that is effectively two-dimensional.  I have two parallel
plates, which are each parallel to the xy plane.  Thus, I have what I think
one would call "slab geometry."

Since my system is clearly not isotropic, it does not make sense to compute
radial distribution functions in the standard way (i.e., considering a
sphere around the atom/molecule of interest).  Do you know if it is possible
to compute radial distribution functions in cylindrical coordinates?

In the documentation for g_rdf, there is a switch "-[no]xy" which by default
is set to "no."  This switch is described as "Use only the x and y
components of the distance."  Am I correct in understanding that "-xy" is
effectively computing the RDF in cylindrical coordinates?  I guess this
would mean that, for every atom/molecule of interest, you would consider a
cylinder whose axis is parallel to the z axis (and the z axis is
perpendicular to the plates, which are parallel to the xy plane). 

Basically, my question is the same as that asked here:
http://lists.gromacs.org/pipermail/gmx-users/2009-April/041475.html ,
although it doesn't seem that that post was resolved.  

Thank you!

Andrew DeYoung
Carnegie Mellon University