AMBER Archive (2003)

Subject: Re: AMBER: Problems restarting with sander

From: raist_at_mmb.pcb.ub.es
Date: Fri Nov 14 2003 - 19:14:42 CST

The failure was on the very first step:

APPROXIMATING switch and d/dx switch using CUBIC SPLINE INTERPOLATION
 using 5000.0 points per unit in tabled values
 TESTING RELATIVE ERROR over r ranging from 0.0 to cutoff
| CHECK switch(x): max rel err = 0.3251E-14 at 2.581700
| CHECK d/dx switch(x): max rel err = 0.8156E-11 at 2.782960
 ---------------------------------------------------
| Local SIZE OF NONBOND LIST = 2513259
| TOTAL SIZE OF NONBOND LIST = 2513259
 vlimit exceeded for step 34; vmax = 24.9785933

     Coordinate resetting (SHAKE) cannot be accomplished,
     deviation is too large
     NITER, NIT, LL, I and J are : 0 0 37 120 121

     Note: This is usually a symptom of some deeper
     problem with the energetics of the system.

The error was betwwen a C and a H with bond distance =1.09 I couldn't see
anything wrong when visualizing the structur
e
I will try to run the simulation incresing taup to see what happens

Many thanks

> On Fri, Nov 14, 2003, raist_at_mmb.pcb.ub.es wrote:
>
> > I'm running a MD on a 8-MER DNA using amber7. I restart the simulation
> > with coordinates and velocities from the previous run every 200ps. On one
> > of this restartings the MD died because of a large deviation between two
> > atoms(common SHAKE error)
>
> The imaging one would do in ptraj is only for visualization; it should have
> nothing to do with SHAKE. In any event, you would not have a SHAKE constraint
> between two strands (I would imagine).
>
> I have no idea why what you did had an effect. You didn't say how soon after
> the restart the SHAKE failure occurred in the run that failed. Did it happen
> on the very first step, or later on? SHAKE failures during runs are chaotic
> events, so *any* change in starting coordinates could lead to a new trajectory
> that might not exhibit the problem. This could be an explanation for the
> behavior you see.
>
> Try a larger value of taup (the default is really too small). Try setting
> ntc=ntf=2, not 3. If neither of those helps, you may need to use a shorter
> time step.
>
> ..good luck...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
> ==================================================================
> -----------------------------------------------------------------------
> The AMBER Mail Reflector
> To post, send mail to amber_at_scripps.edu
> To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu
>
>

-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu