[gmx-users] Distribution of z-coordinates for a particular index group

[Andrew DeYoung]

Hi,

My system is in the slab geometry; I have applied periodic boundary
conditions along x and y, but none along z.

I frequently use the handy tool g_potential (using the -oc switch) to
compute the (average) charge density (i.e., charge distribution or a
"histogram of charges") along z for a particular group specified by an index
file entry.  Since the partial charges are discretely assigned to individual
atoms, this means that the -oc switch of g_potential effectively gives me a
sort of distribution of the atoms along z, although the units are charge
density, of course.

My question is, what if I have atoms with zero partial charge?  Of course,
such atoms do not contribute to the charge density.  But what if I would
still like to compute some sort of distribution of these atoms along z?

One idea is to use g_traj to print all of the z-coordinates of a particular
group over time to an .xvg file. Then, using a scripting language, I could
"bin" the z-coordinates and then average over time.  But is there already a
Gromacs utility that does this?

It seems that g_density, with the "-dens number" switch, may accomplish
this.  Is it true that the "-dens number" switch will lead to computation of
the number density of _each_ atom which is specified in the index file
entry?

Thank you for your time!

Andrew DeYoung
Carnegie Mellon University