AMBER Archive (2005)

Subject: Re: AMBER: nmode minimization and normal mode calculation

From: David A. Case (case_at_scripps.edu)
Date: Tue Aug 30 2005 - 09:38:14 CDT

On Tue, Aug 30, 2005, Bing Xiong wrote:
>
> I first use the nmode minimize my protein to the drms < 1*10**(-3),
> then this saved coordinates were used as input for normal mode
> calculation. But nmode complain the drms > 1*10**(-3). So I am
> wondering is there a way let nmode first minimize several steps then
> change to do the normal mode calculation. Or anyone has the
> experience to deal with this situation? Thanks so much for your help!!
>

If the energy in nmode is *almost* the same as in sander (just different
because of round-off error) you can just increase the tolerance that is looks
for (drms) and see what happens. Or, you can save the coordinates in
unformatted form before transferring to nmode.

But you should minimize to below 10**-3 if you want to do normal modes. Try
running some newton-raphson minimization (ntrun=4) first.

Also, it is easy to have the settings different in nmode and sander....this
will show up as very different energies in the two programs.

...good luck...dac

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