Dear Andrew,
> I did molecular dynamics with zinc as ion and with weak restrains on
> the loop which binds zinc.
> "Keep protein fixed with weak restraints 10.0"
> Zinc is coordinated by two cysteines and one histidine. After 50
> picoseconds zinc atom starts to migrate into the solution from
> cysteins and His, while other structure is not yet much affected.
> It is not in the binding site, so for now as the next step I
> consider to fix it with more strict restrains, what restrain values
> can you recommend?
I am not surprised by what you describe: I would derive charges for
the entire complex (i.e. with bonds between the metallic center & the
ligands) and generate a force field library for your entire complex.
This means time-consuming QM jobs and NOT avoiding the Zn-complex
parametrization ;-)
You can use R.E.D. Server Development for that .
http://q4md-forcefieldtools.org/REDS-Development/
R.E.D. Server Development/R.E.D. IV contains by now a first version of
a new QM algorithm that perform many checking on the optimized
geometry for metallic complexes (a limitation here is that the cpu
time is greatly increased...).
Once your complex is characterized by QM and parametrized, you can run
MD simulations to get a stable complex.
Once the 3D-structure of your complex fully characterized by MD you
might remove the bonds and used a Zn atom with a +2 value. Doing the
opposite way or directly using Zn2+ will likely lead to an unstable
complex - we did it ;-)
We will release a iron complex in R.E.DD.B. (the review is done; see
the projects under review in R.E.DD.B.) following the strategy
described in this email soon.
--
If you really want to avoid QM jobs, you could use NMR restraints to
force your complex to remain stable for 'some' MD time, and then
remove these NMR restraints. But, here where is the proof that this
MD-complex is valid?
Thus, these time-consuming QM jobs allows you to rigorously
characterize the entire complex...
regards, Francois
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Thu Sep 15 2011 - 00:30:03 PDT
