On Tue, Jan 15, 2013 at 2:22 PM, Divi <dvenkatlu.gmail.com> wrote:
> Hello all:
>
> I am running MM/PBSA on a homodimer that has three disulfide bonds in
> each monomer and one disulfide bridge between the monomer. I created the
> gas-phase topology files for two monomers (that are same since it is
> homodimer) and complex using tleap.
>
MM/PBSA is an analysis for non-covalent binding events, only.* If I
understand you, you want to calculate the binding free energy between two
monomers that are linked via a disulfide bond --- you cannot use MM/PBSA
for this.*
> Since original solvated complex has disulfide bonds explicitly defined
> (seven total), I used the same information to create gas-phase topologies.
> The MMPBSA calculation ran fine but with the following error:
>
> WARNING: INCONSISTENCIES EXIST WITHIN INTERNAL POTENTIAL TERMS. THE
> VALIDITY OF THESE RESULTS ARE HIGHLY QUESTIONABLE
>
This check was added to MMPBSA.py when running with a single trajectory as
a failsafe against bad or inconsistent topology files. In a single
trajectory, the conformations in the bound and unbound states for the two
components are identical, so all bonded potential terms should be identical
in both the bound and free states. Ergo, they should cancel completely.
Since there is a bond in your complex that is not in your free states,
this approximation breaks down and you will have non-zero
bond/angle/dihedral delta terms.
Force field parameters are not built to give you an absolute energy for a
particular term (and in fact will not, since a bond has a minimum energy of
0 in a force field that assumes a harmonic potential, but a negative
'absolute' energy).
> If I remove the disulfide bridge information in the leap input file and
> create the gas-phase topology files, the above WARNING goes away, but the
> DeltaG values are very high.
>
Because now the bond that was causing the problems is gone everywhere. But
now you have clashing sulfur atoms, which is equally unphysical.
*There are potentially some questions that you can use MM/PBSA to answer in
this case, but you'll have to use it very carefully (and you cannot use it
the way you've tried to).
HTH,
Jason
--
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 Jan 15 2013 - 13:30:04 PST
