I'm of the opinion that you should use 1. In the GB model, the 'internal'
dielectric is actually the dielectric constant of the medium you transfer
*from*, not really the internal dielectric constant of the protein. What I
mean here is that, in GB, the solvation free energy is estimated by
calculating the energy required to transfer the protein from vacuum/gas
phase (where the dielectric constant is 1) into water (where the dielectric
is 80). The dielectric constant of the protein itself actually cancels in
the derivation.
While there is some rationale to using intdiel != 1, the GB models were
parametrized with intdiel=1 in mind, and I would suggest against deviating
much from that value (in particular, values much higher than 2 are hard to
defend). In particular, I think you would have a hard time defending the
choice of intdiel=10 simply because it gives you answers you expect. You
can use pairwise decomposition to analyze which specific residue-residue
interactions give rise to your unexpected result, but I would be wary of
damping out internal electrostatic interactions to get the behavior you
expect.
HTH,
Jason
On Tue, Jun 12, 2012 at 7:13 AM, Jignesh Patel <jbp087.gmail.com> wrote:
> Hi,
>
> I have performed MD simulations of CCR2 with membrane (POPC). I have also
> estimated per residue decomposition free energy using MM/GBSA method by
> perl script. Residues which shows hydrophobic and hydrogen bond
> interactions with ligand shows good decomposition free energy (about -8 to
> -2 kcal/mol). But, glutamic acid which shows ionic interaction with the
> quaternary nitrogen of ligand gives unfavorable decomposition free energy
> (about 8 kcal/mol). Here I kept external and internal dielectric constant
> values 80 and 1 respectively. Then, internal dielectric constant was
> changed to 10 which gave decomposition free energy of glutamic acid about
> -2.5kcal/mol.
>
> Can anyone tell me what values should I use for internal dielectric
> constant so I can get reasonable decomposition free energies for residues
> showing both hydrophobic and ionic interactions?
>
> Thanks in advance.
>
> With regards,
> Jignesh
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
--
Jason M. Swails
Quantum Theory Project,
University of Florida
Ph.D. Candidate
352-392-4032
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue Jun 12 2012 - 04:30:04 PDT
