On Wed, Oct 26, 2011 at 6:25 PM, Mo Chen <mc842.cornell.edu> wrote:
> Dear Amber users/developers,
> May I ask if the direction of the random forces in SMD simulation could be
> defined somehow?
What random forces are you talking about? The only random forces I know
about implemented in Amber right now are in Langevin dynamics, and that has
nothing to do with SMD. The force induced by the moving umbrella potential
is in no way random. It is deterministic, and analytically calculated from
functional form of the potential (it's just a quadratic equation).
> And are all the variables listed in the example input
> files? Or can I use all the variables introduced in NMR &rst list? Thank
> you
> very much!
>
I don't know what example input files you're looking at, but you should
restrict yourself to defining an umbrella potential. That is, set r1 very
low, to the point that the system will never sample that value or past it,
set r2=r3 so there's no "flat well" potential, set r4 to be very large to
the point that the system will never sample that value or past it, and set
rk2=rk3 so the potential is symmetric. Then you can define the starting
position of the potential and the ending position of the potential (I can't
remember how off the top of my head, it's in the manual), and use nstlim to
control how quickly the coordinate is changed.
That's about the extent of the flexibility you have with SMD (but do you
need any more?). The main issue is calculating free energies (which is an
equilibrium quantity) through these non-equilibrium processes. You'll need
the Jarzynski equation for this, and you need to satisfy the conditions of
this equation for a valid process.
HTH,
Jason
HTH,
Jason
>
> Best,
> Mo Chen
> Graduate student
> Cornell University
> Ithaca, NY 14850
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
--
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 Wed Oct 26 2011 - 21:00:04 PDT
