AMBER Archive (2008)

Subject: Re: AMBER: Building O-Ac modified monosaccharides consistent with Glycam force field parameters

From: Lachele Foley (Lists) (lf.list_at_gmail.com)
Date: Fri Nov 07 2008 - 15:25:18 CST


First, a couple questions: Do you just want to terminate
carbohydrates with O-Ac (that is, at the anomeric C only) or do you
want to substitute various hydroxyls at other positions in the sugar?
If it is the latter, do you need lots of flexibility or are there only
a few substitutions of interest to you? The work required to get
decent (or, at least, consistent) results will depend on what,
exactly, it is that you want to do. It might also be important to
know the intended application in order to know what sort of charge
generation will be sufficient. Generating really good charge sets is
not a trivial task. Generating grossly approximate charges, if that's
all you need, might not be so hard. Do keep in mind, as Karl said,
that the force field is generated as a complete unit; the charges will
affect the interplay between the other parameters as well.

The best way to determine whether your charge set is good, of course,
is to run a simulation that can be compared to one or more
experimental values or to another simulation known to do a really good
job. This, of course, isn't always trivial, either. There isn't
always a strong overlap between what can, in practical terms, be
measured in an experiment and what can, in practical terms, be
calculated.

Regarding the CH3 group... in Glycam, we set the charges on aliphatic
hydrogens to zero. So, the H's will certainly seem to be
"excessively" charged unless your calculation does the same. I see
the output provides you Mulliken charges where the hydrogens' charges
have been absorbed into the nearby atom, which is a similar sort of
thing. As to the carbon, it is probably very worthwhile to calculate
charges for an entire sugar residue with your O-Ac in it's intended
chemical environment before you decide -- and remember that charges
from a single conformation will not reflect an ensemble-averaged
charge set.

There is a description of the method for calculating ensemble-averaged
charges in this paper, if you want to try that: Karl N. Kirschner,
Austin B. Yongye, Sarah M. Tschampel, Jorge González-Outeiriño,
Charlisa R. Daniels, B. Lachele Foley, Robert J. Woods. GLYCAM06: A
generalizable biomolecular force field. Carbohydrates. 2008. Journal
of Computational Chemistry. 29 (4) pages 622 - 655.

If you decide that the charges seem sane enough for your application,
you should just be able to modify the antechamber-generated prep file
with the new charges. If you intend to link your new residue with
other Glycam residues, the un-linked residue charge needs to be set
correctly. Glycam residues are made to be modular, with complementary
partial charges, so that they can be strung together in any
combination.

I'm sorry I can't give an opinion on free QM softare. I only have
experience with a few commercial varieties, so I can't endorse (or not
endorse) any particular package. Here is a page I found after a quick
search that lists several packages, free and commercial:

http://en.wikipedia.org/wiki/Quantum_chemistry_computer_programs

If I seem to have misunderstood any parts of your email, let me know
and I'll see if I can do better.

:-) L

On Fri, Nov 7, 2008 at 10:24 AM, Francois Theillet <ftheille_at_pasteur.fr> wrote:
> Dear all,
>
> I want to add O-Ac on carbohydrate residues. I read the detailed help of
> Karl Kischner http://archive.ambermd.org/200509/0150.html, and followed the
> indications of the tutorial A1 (thank you to all of their authors).
> The "fast and dirty method" would be the following : I have to determine the
> partial atomic charges of the O-Ac in a Met-O-Ac molecule using
> HF/6-31G*//HF/6-31G* RESP 0.010, and then replace correctly the charges in
> the antechamber generated prep file.
>
> However, I do not have any access to Gaussian softwares, and I tried to use
> the free WebMO Demo server.
> Unfortunately, jobs are limited to 60s on this shared resource, and the
> optimization of geometry and partial charges calculation (using a Gaussian
> engine) can be done at the HF/3-21G level of theory only (after a first
> optimisation at the HF/STO-3G level of theory) (and the RESP methodology is
> apparently not used for the charges calculation).
>
> I know that Glycam force field parameters are done at the HF/6-31G*/ level
> of theory and that there is a risk that the partial charges using another
> methodology are not consistent with Glycam parameters.
>
> Thus, I would like to know what would you think if I used this basic WebMO
> demo server output for modifying an antechamber geerated prep file for the
> new residues (apart from the problem of the ensemble of conformations).
>
> (If we compare with the charge parameters of a GlcNAc for example, it seems
> to me that the the Met-atomes (C and H) of the O-Ac are excessively
> charged.)
>
> I attached the ouput of the WebMO calculations.
> I hope that my account and question are not too stupid...
>
> I thank you in advance for your answers,
>
> François Theillet.
>

-- 
:-) Lachele
Lachele Foley
CCRC/UGA
Athens, GA USA
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" (in the *body* of the email)
      to majordomo_at_scripps.edu