[MMTK] Errors in energy evaluation with mg ion

[hinsen at cnrs-orleans.fr]

On 20.07.2004, at 23:28, Eric Zollars wrote:

> Ryan-
>    The last lines of the file ForceFields/Amber/amber_parm94  
> (everything below the line MOD4   RE) are used in the vdW calculation.  
>  Konrad referenced all the values as well.  There does not currently  
> appear to be a Mg.
>
This is indeed the problem: there are no Amber parameters for Mg.

There is no way around obtaining those parameters (Lennard-Jones  
parameters plus partial charge) and I can't help much with that. Once  
you have those parameters, here's how to communicate them to MMTK:

- For the charge, just set
      magnesium_atom.amber_charge = -42.
   in your simulation script. Alternatively, you can add
      amber_charge = -42.
   to the database file for Mg, but that makes sense only if the partial
   charge of Mg is roughly the same in all simulation settings.

- For the Lennard-Jones parameters, you can either hack the Amber
   parameter file (quick and dirty) or write an Amber parameter extension
   file, whose layout is described in the Amber documentation (I can
   provide an example if that helps) and specify it when creating the
   Amber94ForceField object.

Konrad.
--
------------------------------------------------------------------------ 
-------
Konrad Hinsen
Laboratoire Leon Brillouin, CEA Saclay,
91191 Gif-sur-Yvette Cedex, France
Tel.: +33-1 69 08 79 25
Fax: +33-1 69 08 82 61
E-Mail: hinsen at llb.saclay.cea.fr
------------------------------------------------------------------------ 
-------