AMBER Archive (2009)

Subject: Re: [AMBER] AMBER FF and NAMD

From: Nicolas Sapay (nicolas.sapay_at_cermav.cnrs.fr)
Date: Fri Oct 16 2009 - 09:59:53 CDT

stefano_elli_at_libero.it a écrit :
> Hello
>
Hello Stefano
> My name is stefano.
>
> I'm trying to do some MD simulation of heparine like oligosaccharide in explicit water solution.
> I'm using amber force fields with NAMD MD engine.
>
> Ok the right force field to use is GLYCAM but...
> The problem is that heparine like molecule are FULL of charges sulphated group,...and even glycam,...
>
> 1) I have no a gaussian licence, I'll try to calculate RESP charges with gamess.
> My first question is possible to do so?
>
> 2) I'm trying to use gaff force field to have some reference structure.
> I understand that GAFF doesnt' use wdv 10-12 to describe h-bond, but when I start the MD simulation NAMD (2.6 version) read xxx number of h-bond. Surprisingly the job work,...because I read on the NAMD manual that its doesn't like this kind of potential...
>
IIRC, it is just an indication. NAMD doesn't actually treat H-bonds by a
10-12 potential (check the NAMD mailing-list). I made some tests with
NAMD and FF99 and didn't observe any particular issues. To be sure, you
can compare your vdW and electrostatic energies on a single step with
no cutoff. They must be exactly the same between Amber and NAMD.
> So my question is
>
> ->There's a way to describe h-bond with gaff and NAMD?
>
As far as I know, usual force fields (Amber, Charmm, OPLS-AA, ...)
implemented in usual MD engines (Amber, Gromacs, NAMD) do not treat
H-bonds with a specific potential. If you really want a 10-12 potential
on your H-bonds, you will have to play with the code of your favourite
MD engine. I think Gromacs allows user-defined potentials, but I've
never tried that stuff myself.

Nicolas
> Thank you for the answer
>
> Stefano
>
>
> _______________________________________________
> AMBER mailing list
> AMBER_at_ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
>
>



_______________________________________________
AMBER mailing list
AMBER_at_ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber