[gmx-users] compatibility of MOPAC charges with GROMACS ?

[Dallas B. Warren]

> To get more rigorously derived figures, and to calculate 
> charges, I thought of using MOPAC http://www.openmopac.net/index.html
> I imagine that the bond lengths and angles from MOPAC will be 
> reliable. My question is really how compatible the charges 
> are with the GROMACS topologies?  Would you have any qualms 
> about mixing and matching? Surely MOPAC's estimates must be 
> better than human guesstimates or, despite its general 
> wonderful usefulness, PRODRG's estimates?  Are there other 
> programs for this?

What it comes back to is how the forcefield you are using is
constructed.  It will have been done using a particular procedure for
generating the partial charges, L-J interactions, bonding etc to fit
some particular parameters.  If you use a different procedure to
generate yours and try to mix it with the forcefield, then there is an
extremely good chance you will get out garbage.

So what you really need to do, once you have worked out which forcfield
you are going to use, go back to the documentation / papers for that
particular forcefield, see how they generated the parameters, then use
that same procedure for the new parameters / molecules you want to
generated.

Catch ya,

Dr. Dallas Warren
Lecturer
Department of Pharmaceutical Biology and Pharmacology
Victorian College of Pharmacy, Monash University
381 Royal Parade, Parkville VIC 3010
dallas.warren at vcp.monash.edu.au
+61 3 9903 9524
---------------------------------
When the only tool you own is a hammer, every problem begins to resemble
a nail.