On Sun, Sep 25, 2016 at 11:56 AM, Andrew Schaub <aschaub.uci.edu> wrote:
> Good Morning Amber Community,
>
> I was working through the constant pH tutorial (
> ambermd.org/tutorials/advanced/tutorial18/section2.htm), and noticed that
> the heating stage takes 2 nanoseconds (nstlim=1000000, dt=0.002). I'm just
> curious what the rationale is, because I don't believe it's a typo as it
> clearly states in the tutorial, that the first 500,000 steps are to
> linearly increase the temperature:
>
It's personal taste with a hint of pragmatic compromise. Rigorously
correct heating is done infinitely slowly while infinitely fast heating is
done instantly. You need to find the goldilocks number (not too fast, not
too slow) so you get reasonable heating without spending all your
computational resources on a step that will just be discarded in analysis.
I picked 2 ns because on a GPU that is very fast. The simulation heats up
over the first ns and then lets the system "stabilize" during the second,
under restraints to keep the simulation from blowing up. I've seen
reasonable behavior with that combination, so I used it in the tutorial.
But YMMV.
"The *nmropt* option here causes us to vary the target temperature linearly
> from 10K to 300K from step 1 to step 500000 (half the simulation length). "
>
> I am heating and running productions on systems of a larger size (~43 kDa
> (~700 residues)), so I am looking for any speedups I can get. Thanks!
>
The larger your system, the longer you should probably heat to let the bad
contacts relax and the system stabilize. But there's nothing special about
the specifics of the simulation protocol I applied in the tutorial, and if
a shorter heating stage has worked for your system in non-CpHMD
simulations, chances are it would be fine with CpHMD as well.
As the tutorials themselves state, their purpose is to teach you how to use
various protocols -- not necessarily to provide a blueprint of suggested
practices (at the top of each page):
(Note: These tutorials are meant to provide illustrative examples of how to
use the AMBER software suite to carry out simulations that can be run on a
simple workstation in a reasonable period of time. They do not necessarily
provide the optimal choice of parameters or methods for the particular
application area.)
HTH,
Jason
--
Jason M. Swails
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Sep 26 2016 - 09:00:04 PDT
