AMBER: The convergence of energy minimisation

From: Binbin Liu <bmbbl.bmb.leeds.ac.uk>
Date: Tue, 28 Sep 2004 20:10:56 +0100

Dear Amber people,

I am trying to do minimisation on a 3D model with 3 water molecules and 3 ions
in the active site.

The problem is the total energies of the model with waters and ions could not
reach a stable point. Two sets of runs were tried:

1. 500 iterations run, the total energies keep changing to the 500th run. The
RMS at the end is around 2.95

2. 1000 iterations run, the total energies change in a small range about 10
kcal/mol in the first 500 iterations, but between the 500th and 1000
iterations there is a big drop. Moreover at the end, it still not be to the
stable point. The final RMS is about 1.76 something.
 
The parameters I used for the minimisation are basic and simple. They are:
&cntrl
    imin=1,
    maxcyc=1000,
    ncyc=500,
    ntmin=1,
    ntpr=2,
    ntr=0,
    ntb=0,
    cut=12,
&end

I wonder what the realistic standard for the convergence of energy
minimisation and what causes the above things to happen.

Any information is appreciated so much.


Best regards,

Binbin


-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber.scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo.scripps.edu
Received on Tue Sep 28 2004 - 20:53:01 PDT
Custom Search