On Mon, Apr 12, 2004, Holly Freedman wrote:
> We are trying to perform AMBER simulations on hydroxypyridine and
> oxopyridine. We ran Gaussian using ESP to get the charge fitting and
> then got an AMBER prep file and a force field modification for gaff
> using antechamber. We added water, relaxed, thermalized to 298 and then
> equilibrated the structure. We have been leaving the structure fixed at
> the minimum energy structure from Gaussian. However about 70000 steps
> into the md, we started to get messages about vlimit being exceeded and
> also vmax = nan. Would you please have a look at my input and the
> output meesage and give me your advice on how to fix the input?
> Here's the input file for the md run:
>
> &cntrl
> imin = 0,
> ntx = 1, irest = 0, ibelly = 1,
> ntpr = 100, ntwr = 100, iwrap = 1,
> ntwx = 100, ntwe = 5000,
> ntf = 2, ntb = 2, cut = 10,
> nstlim = 100000, dt = .002,
> tempi = 298, temp0 = 298, ntt = 1,
> ntp = 1, taup = 1.0,
> ntc = 2,
> &end
> Moving region
> RES 2 826
> END
> END
>
> Here is the error message in the output file:
>
> NSTEP = 74200 TIME(PS) = 159.400 TEMP(K) = 420.11 PRESS = -154.6
> Etot = -5216.0784 EKtot = 2066.2222 EPtot = -7282.3006
> BOND = 0.0000 ANGLE = 0.0000 DIHED = 0.0000
> 1-4 NB = 0.0000 1-4 EEL = 0.0000 VDWAALS = 1031.1950
> EELEC = -8313.4956 EHBOND = 0.0000 RESTRAINT = 0.0000
> EKCMT = 1040.4340 VIRIAL = 1136.4725 VOLUME = 28777.8501
> Density = 0.8631
> Ewald error estimate: 0.1118E-01
> ------------------------------------------------------------------------------
>
> vlimit exceeded for step 74238; vmax = nan
I don't think it's possible to debug this just from the information you
have, but here are some comments and things to look at:
1. You are setting ntx=1, irest=0, which means that you are basically
"starting over", and not getting any benefit from your initial equilibration
run. Once you have done some equilibration, subsequent runs should use ntx=5,
irest=1.
2. It is possible that using the belly option is causing some problem. It is
this that answers Carlos Simmerling's question about why you have no bond,
angle or dihedral terms -- they are removed from the belly, and everything
(I'm guessing) is water. The Amber 8 Users's Manual adds the following to the
"ibelly" option description: "Note that this option does _not_ provide any
significant speed advantage, and is maintained primarily for backwards
compatibilty with older version of Amber. Most applications should use the
_ntr_ varialbe instead to restrain parts of the system to stay close to some
initial configuration."
But having said this, I don't _think_ belly is causing your problem. But
removing that option would be something to try.
3. You are trying to equilibrate both the temperature and the pressure
(volume) at the same time, which is generally a bad idea. It is certainly
much safer to equilibrate first with ntb=1, ntp=0 to the desired final
temperature, then switch to constant pressure (ntb=2, ntp=1). It is of
concern that your temperature (above) is very high (420K) and the density
is very low (0.86). The problems with the simulation probably began much
earlier than this. Generally, it is safest to equilibrate at dt=0.001,
and only switch to dt=0.002 after equilibration (if at all).
4. Check to see if you have any atoms of type "ho" in your solute (sounds
likely from hydroxypyridine). This hydrogen could have a zero vdW radius,
and that in turn can lead to some bad regions of configuration space where
that hydrogen can get too close to a negatively charged water oxygen atom.
5. Consider posting your prepin file from antechamber. That way, others could
try to do a water simulation and see if they can reproduce the problem.
....good luck...dac
--
==================================================================
David A. Case | e-mail: case.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
==================================================================
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber.scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo.scripps.edu
Received on Tue Apr 13 2004 - 01:53:00 PDT
