there are quite a few studies in the literature now, but perhaps the most
extensive comparison to experimental data (mostly NMR) has been done by Paul
Maragakis and others at the DE Shaw group and I'm not sure if it has been
published, though it has been presented at multiple conferences for over a
year now. They have compared all of the major force fields, including
several versions of each, and 99SB is pretty clearly better than ff03 on a
variety of systems for things like NMR order parameters, scalar couplings
and so on. Their data also shows that 99SB is better than current verions of
other force fields, such as charmm and gromos, but 99SB does indeed have a
few things that need tweaking. My group also has a study in press at Biophys
J. We have been working on the evolution of these parameters ever since
publishing 99SB and hope to have something ready by the end of this year
that will further improve this force field.
On Fri, May 1, 2009 at 4:33 AM, Hannes Loeffler
<hannes.loeffler.stfc.ac.uk>wrote:
> On Thu, 2009-04-30 at 14:36 -0400, Carlos Simmerling wrote:
> > either one could be used in principle, but from the protein perspective I
> > think we have found that 99SB provides a better model. also I think the
> > charge model for 99SB is more compatible with that of glycam, while ff03
> > uses a different charge derivation model. I don't think anyone has done a
> > careful study to determine if this causes any problems.
>
> Just wondering. If you say that 99SB is a better model than ff03 but
> probably no careful study has been carried out yet what is your
> statement based on? I suppose it comes from your personal experience
> but how general would you think that would be?
>
> Hannes.
>
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Wed May 20 2009 - 14:01:21 PDT
