On 18 September 2014 15:59, Jason Swails <jason.swails.gmail.com> wrote:
> On Thu, 2014-09-18 at 15:15 +0300, Thomas Evangelidis wrote:
> > >
> > > MMPBSA.py never accounts for PBC. There is only one GB implementation
> > > that I'm aware of that does (and I don't know how much sense such an
> > > approach makes).
> > >
> > > You may want to look into the membrane capabilities of PBSA. It's
> > > relatively new functionality, but it was designed for studies like this
> > > (it treats the membrane and solvent implicitly by setting up multiple
> > > dielectric regions).
> > >
> > >
> > Thank you. I just had a look. I must define the membrane thickness and
> the
> > cylindrical exclusion region at the center of the protein. I have done
> this
> > in the past with APBS for individual protein models and I spent a lot of
> > time for trial, error and visualization until I find the optimum values,
> > which were still not perfect. I think I could avoid all this if I measure
> > relative free energies of binding for the ligands I have as I would do
> for
> > a water soluble protein. Unless I am missing something, the terms
> > accounting for solvation and desolvation of the transmembrane part of the
> > protein with cancel out.
>
> Can you make a similar argument about cancellation of errors with
> regards to not having periodic images of the membrane?
>
> I believe you are in uncharted waters here with MM/PBSA and MM/GBSA
> analyses, so you will likely have to try and answer most of these
> questions by trying stuff out.
>
>
Does the same apply to alanine scanning to estimate the contribution of the
mutated residue to ligand binding? Do you reckon the membrane slab is
necessary for this type of calculation too?
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Thu Sep 18 2014 - 06:30:05 PDT
