[gmx-users] Topologies and charges for large organic ligands

[Dean Cuebas]

Dear colleagues,

In 53a6, the topology and charges for (NAD+) nicotinamide adenine
dinucleotide are present (listed as NADH, 52 atoms) in polar hydrogen form
(without aromatic hydrogens).

Is there anyone who can tell me how such a molecule was charge group
parameterized? (with sufficient certainty to be included in the ff).

Please understand that I am not asking how to use the NAD topology that's
provided.

I've read that prodrg is a "good starting point" for parametrizing a ligand,
but what is the path to "improving" such a ligand from that starting point?
(the provided topology for NAD+ in the ff shows 17 charge groups with
an average of 3-4 atoms per charge group, whereas prodrg gives only 11
charge groups.)

I'm not concerned about angles and dihedrals, since that is pretty
painless... it's the charge groups that I'm concerned about.

In a nutshell, who and how was the final NAD+ parameters decided upon??
Does anyone know this??  Is it simply chemical intuition of estimated
charges that reproduce the dipole of the moiety (like an adenine ring), and
the charge groups contain the minimum number of atoms that reflect this?

Is there an algorithm to generating "improved" charge groups?

I ask these questions because my ligands are large organics... MW 800 and
greater.

Thanks for any and all help in this regard.

-- 
Dr. Dean Cuebas, Associate Prof of Chemistry
deancuebas at smsu.edu, Ph 417-836-8567 FAX 417-836-5507
Dept. of Chemistry, Southwest Missouri State University
Springfield, Missouri 65804