AMBER Archive (2004)

Subject: Re: AMBER: atoms out of bounds

From: Carlos Simmerling (carlos_at_ilion.bio.sunysb.edu)
Date: Thu Oct 07 2004 - 07:17:15 CDT


Your system is nonperiodic. I think you're just seeing the problems that you
can have with this kind of method. if you know where the ion should be, you
might want to use restraints to keep it close to the protein. if you do
not know,
then maybe your initial guess was not good. equilibrating the positions
of ions
is not trivial, and doing it in a nonperiodic system can have problems
because
there is nothing to keep the ions close by the protein after it
disocciates the
first time (as you have found). I think that the approximations of using a
distance-dependent dielectric are large enough to make the use of explicit
ions pointless unless you know that a particular ion is needed in the
structure.

===================================================================
Carlos L. Simmerling, Ph.D.
Associate Professor Phone: (631) 632-1336
Center for Structural Biology Fax: (631) 632-1555
Stony Brook University Web: http://comp.chem.sunysb.edu/carlos
Stony Brook, NY 11794-5115 E-mail: carlos.simmerling_at_stonybrook.edu
===================================================================

Balvinder Singh wrote:

>I am running a dynamics calculations using amber7 on a protein neutrilized
>by one ion. I have minimized the system with DRMS 0.00001 with distance
>dependent dielectric. I have equilibrated the system for 500 ps and
>raising the temperature (from 0K to 300K) in smaller steps of 30K/50 ps.
>Visual inspection of protein at the end of equilibration shows that IB has
>moved a bit far away from the protein (nearest atom of prtein to IB is 64
>A distance). While subjecting the system to production run, following
>error occured.
>
>-----------------
>Frac coord min, max: -8.20798303E-07 0.879446721
> The system has extended beyond
> the extent of the virtual box.
> Restarting sander will recalculate
> a new virtual box with 30 Angstroms
> extra on each side, if there is a
> restart file for this configuration.
> EWALD BOMB in subroutine Routine: map_coords (ew_force.f)
> Atom out of bounds. If a restart has been written,
> restarting should resolve the error
>-------------
>
>The above error also appears to be related to that only whereby the IB is
>moving further away from the protein and resulting in atoms out of bounds.
>The protein is intact. I am sure that restarting it is not going to solve
>the problem as IB may move further away as simulation progresses.
>
>What are the possible reasons for IB drifting away? How to overcome this
>problem?
>
>Latest input of equilibration step (It was performed in steps):
>
>&cntrl
> imin=0, irest=1, ntx=5, ntb=0, cut=50,
> ntr=0, tempi=300, temp0=300, ntt=1,
> tautp=1.0, dielc=4,
> ntpr= 500, ntwx=1000, ntwr=1000,
> nstlim=50000,
> &end
> &ewald
> eedmeth=5,
> &end
>
>Input for production run:
>
>&cntrl
> imin=0, irest=1, ntx=5, ntb=0, cut=50,
> ntr=0, tempi=300, temp0=300, ntt=1,
> tautp=0.5, dielc=4
> ntpr= 5000, ntwx=10000, ntwr=10000,
> nstlim=9500000,
> &end
> &ewald
> eedmeth=5,
> &end
>----------------------------
>
>
>-Balvinder
>
>-----------------------------------------------------------------------
>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