Atomdefinition for charged and uncharged Atoms

Konrad Hinsen
Mon, 9 Nov 1998 16:00:14 +0100

> I have some questions beforehand.  First, do the methods in the mmtk
> module make a difference between charged and uncharged atoms?  I assume

No. An uncharged atom is simply an atom with charge 0.

> yes, because otherwise a lot would not make any sense.  But if yes, then
> where do the methods get the information about charged residues
> from(like van-der-Waals radii)?  Do I have to have a seperate set of
> definition files for all the charged atoms, like one file for Fe, one
> File for Fe2+ etc?  What would a definition file for a charged atom look

The responsibility for dealing with charges lies with the force
fields. Assuming that you are not planning to implement a new force
field, you will use the Amber94 implementation in MMTK. This force
field takes charge information from an attribute "amber_charge", which
can be defined at different levels (atom, group, molecule, complex).

The easiest way to define a charge is the atomic level: you just write
"amber_charge = 0.42" (or whatever) into the atom definition file.
This is rarely done in practice because it is not useful; charges
depend not only on the atom type, but also on the chemical
environment. Therefore charges are typically defined at a higher
level, i.e. group or molecule. There the charge definition takes
the form of a dictionary. For example, the charges for water are
defined in the file Molecules/water by the line

amber_charge = {O: -0.83400, H1: 0.41700, H2: 0.41700}

A heme group is a complex of an organic molecule and an iron atom.
So you should define:

1) An atom species "fe", in the directory Atoms, containing all
   general data about iron (weight etc.)
2) The organic molecule (I don't even know its name!) in a file
   in the directory Molecules.
3) The heme complex in a file in the directory Complexes.

You could then define charges for the uncomplexed molecule (which
after all exists in an iron-free state) in the molecule definition
file, and the charges for both the molecule and the iron in the
complexed state in the definition file of the complex. This works
because definitions at higher level always override definitions
at lower levels.

The remaining question is of course where to get the parameters, but I
don't know. I suspect someone has already worked with hemes and the
Amber force field, but that's not me! In general, parameters for ionic
systems, especially with metal ions, are difficult to obtain;
the standard empirical force fields don't describe such systems well.

> I guess, somebody else must have had a similar problem before and
> probably posted it.  Is there a way to read old posts from this mailing
> list?

Good question. Everything is archived, but I don't really know how to
get at the archive! I'll have to ask the system administrators on the
machine that is handling the mailing list. Anyway, your question has
not been asked here before!

Konrad Hinsen                            | E-Mail:
Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-
Rue Charles Sadron                       | Fax:  +33-
45071 Orleans Cedex 2                    | Deutsch/Esperanto/English/
France                                   | Nederlands/Francais