Dear Bernhard,
First we have to keep in mind that use of dielectric constants higher
than 1 is only an approximate way to incorporate polarization into
binding since the calculations are done with additive force fields
that are supposed to incorporate polarization into their parameters
during calibration. Nevertheless, the approximation has been working
most of the times, i.e. agreement with experiment is better when the
dielectric constant is set to be higher than 1.
A cleaner approach would be to use polarizable force fields for
binding calculations, but this is still an active research topic.
Thus, it would be very hard to understand why it fails in a specific
case if the calculation is set up correctly at every step. However, in
general, a higher dielectric constant should reduce the contribution
of electrostatic interactions (both Coulombic and solvation
electrostatic, i.e. PB/GB) in the final binding free energies.
All the best,
Ray
On Tue, Jun 7, 2011 at 4:29 AM, Bernhard Poll
<poll.chemie.uni-hamburg.de> wrote:
> Hi everybody,
>
> I recently did some MM-GBSA/PBSA with different dielectric constants. Everything looks fine except one complex which includes a ligand with some carboxylate group (which has to be there). The calculated energy was extremely high. I checked the input- and output-files and everything looks fine there but the energy value seems strange to me. So are MM-GBSA/PBSA calculations with different dielectric constants of charged molecules possible or are they problematic?
>
> Greetings
>
> Bernhard
>
>
>
>
>
> _______________________________________________
> 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 Tue Jun 07 2011 - 11:00:02 PDT
