Dear Prof Case,
Many thanks for reply. I want to use amber coordinates and prmtop for
chemshell, where use of non PBC is recommended. I tried several times using
ntb=0, but I am stuck at heating dynamics due to SHAKE error while same job
runs fine with PBC. Even I am unable to find more detail of explicit water
simulation without water box in either manual or tutorial. I followed the
previous post
http://archive.ambermd.org/201402/0352.html, but it is not
applicable as I am using no solvent box. That's why I am trying to use PBC
trajectories for non PBC use.
In addition, use of Gaussian interface for QM/MM is not optimizing the
QM part, sander writes the energy of QM part before proper convergence and
there is no manual control of 'Force' keyword added by sander in Gaussian
route section. The Force keyword added by sander is forcing Gaussian for a
single calculation only and does not wait till proper convergence.
Therefore I have no more option other than to use Chemshell for QM/MM
optimization that prefers non PBC conditions.
Regards
Kshatresh
On Sat, Jun 14, 2014 at 11:51 PM, David A Case <case.biomaps.rutgers.edu>
wrote:
> On Sat, Jun 14, 2014, Kshatresh Dutta Dubey wrote:
> >
> > I have a simulated a protein ligand complex using PBC, but for the
> > purpose of QM/MM calculations I want to remove the box info from the
> > trajectory 'without striping water'.
>
> I'm not sure why one would want to do this. The same arguments that lead
> to the use of PBC for explicitly solvated MM calculations apply with equal
> force for QM/MM simulations.
>
> You *can* get rid of box information by setting IFBOX=0 in the prmtop
> header, and setting ntb=0 in your sander input file. But then the MM
> energies will be quite different, and you will no longer be "equilibrated"
> (to whateer extent you were before embarking on QM/MM.)
>
> If you are convinced that you need a non-PBC environment for QM/MM, I'd
> recommend doing the MM part the same way.
>
> ....dac
>
>
> _______________________________________________
> 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 Sat Jun 14 2014 - 15:00:02 PDT
