Hi Rilei,
> When I set up dt=0.001, everything goes well. However, when I try
> dt=0.002, some bugs appeared! I even found someone even asked such
> question in the mail-list, but I still found it could not solve my
> problem. Can anyone give me some help?
These are not bugs but rather something telling you there is something wrong
with your simulation.
A few comments on your input file:
> &cntrl
> ntx=1,
> irest=0,
> imin=0,
> ntpr=10,
> ntwx=10,
> ntwr=50,
> nstlim=5000,
> dt=0.002,
> ntt=3,
> gamma_ln = 1.0,
> temp0=300,
> tempi=300,
> tautp=1,
tautp is meaningless with ntt=3.
> igb=0,
> ntb=0,
You are running a vacuum simulation with no solvent whatsoever. I very much
doubt your system is stable in a vacuum at 300K. A lot of biological
molecules decompose under such conditions. I would suggest running with
either implicit or explicit salvation. Unless you are specifically trying to
reproduce vacuum results.
> cut=15,
You should NOT use a cut off with non-periodic calculations. Set this to
99999.0
> ntr=1,
> fcap=10.0,
Are you running a solvent cap? If so, why? fcap is only meaningful in this
situation.
> ivcap=0,
> tol=0.000001,
You set the shake tolerance but nowhere do you turn on shake. Your default
if NTF=1 and NTC=1. This means your hydrogens are free to move. VERY FAST.
Faster than the nyquist limit for a 2fs time step. This is the cause of your
vlimit failures, they are because of integration errors caused by using a
timestep too long for the simulation parameters. Try turning on shake with
NTF=2 and NTC=2.
> ntr=1, restraint_wt=10.0, restraintmask=':1-90, 93-140, 151-180, 196-
> 234, 242-256, 265-278, 283-310, 322-360, 371-410',
> &end
A restraint_wt of 10.0 is probably a little big, reduce this to between 1.0
and 5.0. You should also check that you are providing a reasonable reference
structure, and not one that has massively different coordinates to your
inpcrd.
All the best
Ross
/\
\/
|\oss Walker
| Assistant Research Professor |
| San Diego Supercomputer Center |
| Tel: +1 858 822 0854 | EMail:- ross.rosswalker.co.uk |
|
http://www.rosswalker.co.uk |
http://www.wmd-lab.org/ |
Note: Electronic Mail is not secure, has no guarantee of delivery, may not
be read every day, and should not be used for urgent or sensitive issues.
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Apr 09 2010 - 12:30:05 PDT
