AMBER Archive (2008)

Subject: RE: AMBER: initial command?

From: Steve Spronk (spronk_at_umich.edu)
Date: Fri Feb 29 2008 - 08:17:38 CST

> -----Original Message-----
> From: owner-amber_at_scripps.edu [mailto:owner-amber_at_scripps.edu] On Behalf
> Of Bill Ross
> Sent: Wednesday, February 27, 2008 1:56 PM
> To: amber_at_scripps.edu
> Subject: RE: AMBER: initial command?
>
> > I know that the
> > manual (at least for Amber 8, my version) says that CONECT records are
> > ignored, but the manual is not correct, in my experience.
>
> Have you ever tried to connect two atoms that weren't connected
> in the residue template, and verified the presence of the bond
> in xleap?
>
> Bill

Yes, although I didn't actually use xleap--I use tleap. Perhaps using tleap
or xleap makes a difference in this matter.

The specific experience that I had is that I had a protein with disulfide
bonds, just like the original poster. I had CONECT records in the PDB file,
but because of the manual I thought they'd be ignored when I read the file
into tleap. So I used the "bond" command to manually place the disulfide
bonds. But I kept having problems when I tried to write out the prmtop.
Sorry, I can't remember the specific error.

So I tried again, starting tleap over, but the second time I took a look at
the description of the SG atoms of my CYX residues, and wouldn't you know
it, the disulfide bond already existed. Tleap had created it just from the
information in the PDB file.

I also ran a little test by deleting the CONECT lines from the PDB file. In
all other respects the file was identical. When I read this file into
tleap, the bond was not immediately detected; I had to add it manually.

In either case--having CONECT records in the file and not using the bond
command, or not having CONECT records and using the bond command--the prmtop
file that was written out was identical.

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

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