Thank you, I am using igb=6 now. The minimization is still breaking the
bonds, but the funny thing is that equilibrium runs in the NVT and NVE do
run with no problems! My initial conformation seems to have a good enough
initial geometry to bypass the minimization step.
On Mon, Dec 30, 2013 at 3:32 PM, Jason Swails <jason.swails.gmail.com>wrote:
> On Mon, Dec 30, 2013 at 2:36 PM, Jose Borreguero <borreguero.gmail.com
> >wrote:
>
> > Dear AMBER users,
> >
> > I have the following input file intended for minimization of a small
> > molecule in vacuum. I want to do the minimization with full
> electrostatics
> > but without the Ewald summation, and no boundary conditions. The molecule
> > is small enough so that the electrostatics cutoff (12A) encompasses the
> > whole molecule.
> >
> > Minimization with sander, no boundary conditions
> > &cntrl
> > imin=1, ntmin=0, !invoke full conjugate gradient minimization for X
> > steps
> > maxcyc=20000, !minimization for X steps (20000 is a good value)
> > nsnb=5, !update of pair list frequency, 25 at equilibrium
> > cut=12.0, !electrostatics cutoff
> > ntc=1, !SHAKE flag, 1:no shake, 2: shake hydrogens, 3: shake all
> > bonds
> > ntf=1, !force evaluation, tipically ntf=ntc, 1:complete,
> > 2:no-bonds-with-H, 3:no-bonded
> > ntb=0, !periodic-bound-cond, 0:no 1:yes
> > ntpr=20, !output energy information to mdout and mdinfo
> > ntwx=100, !output coordinates to mdcrd
> > ntwe=20, !output energies and temperatures to mden
> > ntwr=2000, !output restart file
> >
>
> Try setting igb=6. That option gives gas-phase energies.
>
>
> > &end
> > END
> > END
> >
> > Can you please let me know if you find something wrong with this input
> > file? My minimization is breaking the bonds :(
> >
>
> Minimization should not break bonds. It could be that bonds that you think
> are present are not present or that the force constant is too weak. You
> can inspect your topology file with ParmEd and use various print* commands
> (like printBonds, printAngles, etc.) to print out details about the
> parameters.
>
> Good luck,
> Jason
>
> --
> Jason M. Swails
> BioMaPS,
> Rutgers University
> Postdoctoral Researcher
> _______________________________________________
> 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 Mon Dec 30 2013 - 13:30:03 PST
