On Tue, Apr 02, 2013, dbaogen wrote:
>
> Binding free energy between ATP (GTP) and protein is
> calculated from the extracted snapshots of MD trajectory
> using AMBER and CHARMM force field, respectively. But their
> binding free energy order is different. Using AMBER force
> field, the order of binding free energy for ATP and GTP is
> consistent with experiments, although absolute binding energy
> has a large deviation from the experimental data. If CHARMM
> force field is used, the order is opposite which contradicts
> with the experimental results.
I personally would be wary of drawing conclusions about relative binding
affinities when the absolute binding energies are far from correct. The
virtue of the MM-PBSA model lies in its ability to provide a rationalization
for binding affinities and a framework for discussing various physical
contributions to binding. It is often the case that its accuracy is not great
enough to correctly predict relative binding energies with very much
reliability.
You might want to consider why you are not getting good absolute binding
energies. Are you leaving out some entropy terms? Further, binding
of negative ions like ATP or GTP is often accompanied by simultaneous
recruitment of metal ions or protons. Of course, changes in the force
field may also be important, since solvation/desolvation of ions depends
very sensitively on the dielectric radii chosen to define the boundary
between high and low dielectric regions.
To be sure, in comparing ATP to GTP, a lot of fairly large contributions to
binding may nearly cancel, and you might want to look carefully at
contributions to the binding energies specific to the A and G bases and their
respective interactions.
...good luck....dac
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue Apr 02 2013 - 05:00:05 PDT
