AMBER Archive (2008)

Subject: Re: AMBER: for NMR refinement

From: David A. Case (case_at_scripps.edu)
Date: Fri Feb 29 2008 - 12:53:59 CST

On Thu, Feb 28, 2008, Wenyong Tong wrote:
>
> Basically, these two annealing protocol are the same as posted here before except the following weight changes that I mistakenly typed.
>
> TAUTP 0 50000 0.200000 0.200000 0 0
> TAUTP 50001 20000 4.000000 2.000000 0 0
> Warning: ISTEP1 >= ISTEP2
> TAUTP 20001 88000 1.000000 1.000000 0 0
> TAUTP 88001 100000 0.500000 0.500000 0 0
> TAUTP 100001 200000 0.010000 0.050000 0 0
> REST 0 500000 0.100000 1.000000 0 0
> REST 500001 200000 1.000000 1.000000 0 0
> Warning: ISTEP1 >= ISTEP2
>
> Although I got warnings, the annealing still works and result is even
> better than correct one from the aspects of ETOT(-50 vs. 25) and total
> distance penalty( 70 vs. 120).
>
> I was wondering how AMBER recognizes these weight changes with
> warnings and works and how I should correct it.

You'll have to trace through the code yourself to see what happens when
istep1 > istep2. I suppose these warnings should really be errors, since
how it "really" works could be very strange in such circumstances.

...dac

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