[gmx-users] Re: Adding Partial Charges for Small Molecules

[Alan]

Dear Nancy,

Just completing what Justin did, see below.

On Mon, Jun 14, 2010 at 03:10, <gmx-users-request at gromacs.org> wrote:

> Nancy wrote:
> > Hi All,
> >
> > I am trying to add partial charges to nicotinamide adenine dinucleotide
> > (NAD+), for input to molecular simulations.  I have been using the UCSF
> > Chimera program to do this.  The total charge on this molecule should be
> > -1 (assuming the ionization of NAD+ at pH ~7.0, see attached figure).
> > When I add Gasteiger charges, the total charge sums up incorrectly to 0,
> > even though the net charge specified in the drop-down box is -1.
> > However, using the "AM1-BCC" method, the net charge correctly sums up to
> > -1.  The charge model used is "Amber ff99SB".  I have also noticed that
> > the partial charges on individual atoms assigned when using the
>
> Sounds like a known issue with Chimera...
>
> http://www.cgl.ucsf.edu/pipermail/chimera-users/2007-August/001732.html
>
> > "AM1-BCC" method are generally of greater magnitude compared to charges
> > assigned using the Gasteiger method.
> >
> > What is a good way to add correct partial charges to a small molecule?
> >
>
> The best way is to follow the original methodology of the force field
> itself.
> The antechamber program of Amber is probably a good place to start to get
> Amber-compatible topologies.  Also have a look here:
>
> http://www.pharmacy.manchester.ac.uk/bryce/amber#cof
>
> All of the above will require conversion to Gromacs-compatible formats, but
> there are scripts on the User Contribution site that might be suitable.
>

Nancy, definitely, I wouldn't use Gasteiger here unless you're doing a
comparison test.
If you want something the closest to what AmberFF were done with the least
hassle, look at

http://q4md-forcefieldtools.org/REDS/

It simply worth the time reading and registering.

Then, if I may suggest, you can try ACPYPE to generate your topology
parameters in Gromacs format. ACPYPE indeed uses Antechamber, but as the
link for the bug above, it may have the same issue with Gasteiger. However,
as I said, I would use am1bcc, which is straightforward with ACPYPE, and so
far I've been happy with the results, even comparing with REDS.

Best regards,

Alan


> -Justin
>
> > The mol2 structure code (with incorrect Gasteiger charges) is written
> > below (and attached as NAD.mol2 with corresponding PNG structure image).
> >
> > Thanks in advance,
> > Nancy
>



-- 
Alan Wilter S. da Silva, D.Sc. - CCPN Research Associate
Department of Biochemistry, University of Cambridge.
80 Tennis Court Road, Cambridge CB2 1GA, UK.
>>http://www.bio.cam.ac.uk/~awd28<<
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20100614/576d8cf3/attachment.html>