AMBER Archive (2003)

Subject: Re: AMBER: gibbs pmf calculation

From: Lishan Yao (yaolisha_at_pilot.msu.edu)
Date: Wed Aug 27 2003 - 10:34:02 CDT

Hi:
   It seems that gibbs works now.But then it stops quickly due to
coordinate resetting problem. When I look at the restraint energy, it is
a huge number. When I read the manual gibbs restraint part, there are
only bond, angle, dihedral restraints. If I want to restraint any two
atoms, could I just use the bond restraint? I think gibbs might assume a
bond between the two atoms I choose.

Best,
Lishan

NSTEP = 3 TIME(PS) = 0.003 TEMP(K) = 1094.743 PRESS
= 0.0
0
 Etotal = 16531.23820 Kinetic = 42103.92787 Potential =
-25572.68967
 Bond = 4739.74158 Angle = 9662.84700 Dihed =
6638.87777
 1-4 VdW = 3734.78889 1-4 Elec= 36369.29540 Van d Waal =
-7699.33309
 Elect. = -79018.90723 H-bond = 0.00000 Constraint =
0.00000
 Ekcmt = 0.00000 VirT = 0.00000 Volume =
0.00000

 Internal UMBRELLA restraints defined in Gibbs (not inc. in above):
 Bond = 46576.10195 Angle = 0.00000 Torsion= 0.00000

     COORDINATE RESETTING CANNOT BE ACCOMPLISHED,
     DEVIATION IS TOO LARGE
     NITER, NIT, LL, I AND J ARE : 0 1 574 1149 1151

On Mon, 2003-08-25 at 19:07, David A. Case wrote:
> On Mon, Aug 25, 2003, Lishan Yao wrote:
>
> > I fixed the problem by changing my input file. But then the new
> > problem comes:
> >
> > Total accumulated time: 2.53 (seconds) / 0.00 (hours)
> > %GIBBS-I-NEWRUN, Begining NRUN = 1
> > ERROR from RESNBA: Number of non-bonded pairs exceeds the maximum
> > allowable ( 2790891). Increase integer storage.
>
> You must edit the top of gib.f to increase the "memmax" parameter, then
> re-compile.
>
> If you are not experienced with gibbs, I would recommend trying a calculation
> on a smaller problem first, to make sure you understand how the program
> (which is a difficult one) works.
>
> ..good luck...dac 

-- 
Lishan Yao <yaolisha_at_pilot.msu.edu>
MSU

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