On Wed, Oct 03, 2001, Filip Lankas wrote:
>
> tautp - time constant for heat bath coupling for the SOLUTE
> tauts - the same for the SOLVENT
> taup - pressure relaxation time
>
> In A5 the default for tautp and tauts is 0.2 ps but the recommended value
> FOR TAUTP is 0.5-5 ps - no hint for tauts. In A6 the default of tautp has
> been changed to 1.0 ps while tauts seemingly doesn't exist as an input
> parameter (at least I don't see it in the text manual on the web).
There is no distinction in sander6 between solute and solvent: tautp is
used for both.
At some points, recommended and default values of various parameters have
diverged, mostly in a (misguided?) attempt to maintain backwards
compatibility for dusty input decks. We are now much more inclined to
set default values to recommended ones.
>
> This confuses me a bit. Can anybody comment on this? Which values would
> you use? How sensitive is the choice?
Generally, larger values are better if you want to disturb Newton's
equations as little as possible. Shorter values are useful for changing
from one temperature to another, or for doing simulated annealing, etc.
Values of 0.2 are probably OK, but they disturb the dynamics somewhat: it
depends on how much you are interested in time scales, and how much just
in average structures, etc.
I personally generally equlibrate to a good temperature and box size, using
tautp and taup in the range of 0.5 to 1.0, then turn off both temperature
and pressure coupling for production runs. This allows me to interpret
dynamical behavior, and to monitor conservation of energy as a crude
accuracy criterion. Of course, this assumes that the conformation is pretty
stable -- if the molecule changes shape, you should not assume a fixed box
size.
...hope this helps...dac
--
==================================================================
David A. Case | e-mail: case_at_scripps.edu
Dept. of Molecular Biology, TPC15 | fax: +1-858-784-8896
The Scripps Research Institute | phone: +1-858-784-9768
10550 N. Torrey Pines Rd. | home page:
La Jolla CA 92037 USA | http://www.scripps.edu/case
==================================================================
Received on Wed Oct 03 2001 - 08:57:33 PDT
