Hi all,
Minimization by the BFGS methodology generally do not converge in the
closest local minimum. The BFGS methods are implemented in Gromacs and
Arguslab. Both software's
do not have the AMBER FF implemented however (may be Arguslab have the AMBER
FF).
Hope this helps,
Regards,
Carlos Javier
2010/2/25 Jason Swails <jason.swails.gmail.com>
> 2010/2/25 Shaandar Nyamtulga <nyam100.hotmail.com>:
> >
> > I changed the sequence like this {NALA ALA ALA ALA CALA}. And added
> minimization cycle 10 times (2000). No success. And when I see original
> structure from xleap it is very strained structure, otherwise backbone is a
> line. At least Amber should be able to convert this linear backbone into a
> helix after EM. Thank you
>
> 2000 is still a small number of steps, but the point remains that the
> sander minimizer will converge only to the closest local minimum.
> While the linear conformation is certainly not the most stable
> structure for a poly-alanine, it is not *strained* per se, since the
> bonds/angles are in their equilibrium positions.
>
> If you want to see your alanine 5-mer fold into a helix, you'll have
> to run some form molecular dynamics (i.e. simulated annealing can be
> shown to give the global minimum structure if done *correctly*).
>
> You will not see what you're trying to see just doing a minimization,
> though.
>
> Good luck!
> Jason
>
> --
> ---------------------------------------
> Jason M. Swails
> Quantum Theory Project,
> University of Florida
> Ph.D. Graduate Student
> 352-392-4032
>
> _______________________________________________
> 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 Wed Feb 24 2010 - 23:30:04 PST
