AMBER Archive (2006)

Subject: Re: AMBER: Dynamics and Restraint problem

From: David A. Case (case_at_scripps.edu)
Date: Thu Jul 27 2006 - 18:08:16 CDT


On Mon, Jul 17, 2006, Andrew Box wrote:
>
> I have an error with using noe restraints with Amber 8.0. When i construct
> the constraint file, all looks fine, and when i run the file in a dynamics
> run, the runs proceeds to completion. But when i check the output file,
> the noe violations are have the incorrect atom names. This problem only
> occurs in Amber 8.0 (which is the version i am currently running, and is
> also available on our local super computer). When i perform the runs in
> Amber 7.0, i had no problems, and i also tried to use the file i generated
> in Amber 7.0, and i get the same result.
>
> Have any changes been made to how constraint file are generated or read
> since Amber 7?
>
> I have included with this the:
> map file - map.AMBER3
> the pdb file - b2.pdb
> 7 column constraint file (before i ran it through makeDIST_RST) - DIST4
> output file: md5.out
> restraint file - RST2 (Amber *) and RST2_old (Amber 7)
>

It's hard to believe that RST2_old actually came from the DIST4 input file:
the comments in the RST2_old file are just copied over from the entries in the
input file, but the comments in RST2_old are completely different from those
in the DIST4 file you provided. (For example, RST2_old refers to residue
23 as a BEN, but there is no residue 23 in the b2.pdb file.)

Since you then used RST2_old as input to create the md5.out file, it's not
surprising that the atom id's are all wrong.

I suspect(?) cockpit error of some sort when you created RST2_old: it doesn't
contain the information it should, and it looks like it was created from some
input file other than DIST4. What happens if you use RST2 (which looks
OK as far as I can tell) as input to sander?

...hope this helps...dac

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