Re: [Open Babel] Electrostatic Potential Map Calculations

[J.B. B. <jb...@ku...>]

Hi Tim,

Thank you for your response to my inquiry.

Do you mean a Van der Waals surface colored by ESP? or do you mean
> contour surfaces of the ESP?
>

Without completely defining my project...
the primary goal is actually not the visualization of ESP, but rather to
obtain the charge value at each point in the mesh for a given molecule or
pair of molecules.  The visualization of ESP will help in explaining my
research to my colleagues and my boss but is ultimately reduced to a
supplementary role.
It is my understanding that the Gasteiger charge is the magnitude of
attraction/repulsion at each particular atom in the molecule, and this can
be generated by OpenBabel, and explained in a previous thread in 2008.

The second image is the output from obprobe, also visualized with avogadro.
>

Can you briefly discuss the differences between obprobe and Gaussian?  Since
both output Gaussian cube files which are a mesh of points around a
molecule, I am unsure about this.  Is one of the differences the point you
mention below?

The obprobe doesn't really generate an ESP cube. Instead, it uses the
> MMFF94 forcefield's non-bonded (Van der Waals + Electrostatic) energy
> terms to "probe" for attraction/repulsion. Later, contour surfaces for
> these energy cubes could be used to find regions of attraction or
> repulsion. Currently, obprobe doesn't change the dielectric constant
> which would be a huge improvement when using it to examine protein
> binding sites for example.


Is there a particular book or journal paper you recommend for learning about
electrostatic potential in molecules?  I have a copy of Smith and William's
Introduction of Drug Design and Action available...


>
> > To recap, my goal is to do something like this:
> > GIVEN: MOL file m1.mol
> > OUTPUT: EPM file m1.epm (or multiple files representing one EPM for each
> > input MOL file)
>

As I mentioned before, the images would be nice, but the mesh values of
point charges (what I referred to as m1.epm) are the end game.  Indeed, I
will be working with hundreds or thousands of compounds, so automatic image
generation and automatic grid point charge calculation are critical for me.

Thanks so much for your help.

J.B. Brown
Kyoto University Institute for Chemical Research