hi,
what you describe seems like a demonstration of the 'flying icecube
phenomenon' or more accurately 'spinning icecube' in your case.
->velocity rescaling (the berendsen method) tends to slowly drain energy
from the high frequency motions and to pump it into the low frequency
modes (translation and/or rotation (if available)).
my suggestions (in no particular order):
change to constant energy at an earlier time.
increase tautp (which should have roughly the same effect).
if it is not an 'icecube - problem' do not worry about the rotation
( unless centrifugal forces disrupt your structure ;-)
if you like fiddling with other peoples' code,
build langevin dynamics into the md-routine, if the heatbath-coupling is
small, it assures a more 'natural' equilibration between your molecule
and the heatbath.
get
ftp://ftp.scripps.edu/pub/case/nab-4.3.1.tar.gz
(dave case's/tom macke's Nucleic acid builder)
if you get it to work with proteins ( i use it for RNA's only)
(it probably works right out of the
box). you can do implicit solvent MD with it. the ff-routines are written
in clear C , they are 'AMBER-compatible' and easy to adapt to your
needs.
have fun
best regards
andreas
)))))
(((((
( O O )
-------oOOO--(_)--OOOo-----------------------------------------------------
o Wolfgang Andreas Svrcek-Seiler
o (godzilla)
svrci_at_tbi.univie.ac.at
.oooO Tel.:01-4277-52733
( ) Oooo.
-------\ (----( )--------------------------------------------------------
\_) ) /
(_/
Received on Mon Dec 04 2000 - 04:32:17 PST
