[gmx-users] do_dssp: main chain

[Florian Haberl]

Hi,

perhaps your .trr has some errors. Have you tried converting it or gmxcheck?

Other solution could be: Transfer your protein residues to an multiple pdb 
file and try do_dssp again. Perhaps your top or ndx doesn't fit.




On Wednesday, 23. May 2007 17:07, Volker Knecht wrote:
> Dear all,
>
> >Message: 4
> >Date: Wed, 23 May 2007 10:16:59 -0400
> >From: Justin Lemkul <jalemkul at vt.edu>
> >Subject: Re: [gmx-users] Re: Re: do_dssp for many peptides
> >To: Discussion list for GROMACS users <gmx-users at gromacs.org>
> >Message-ID: <1179929819.46544cdb08c4f at webmail.vt.edu>
> >Content-Type: text/plain; charset=ISO-8859-1
> >
> >>> do_dssp claims that are 248 residues in my selected group, although 248
> >>> is the number of *atoms* (not of residues) of this peptide. After the
> >>> notion that some temporary file is backed up, do_dssp gets stuck again.
> >>>
> >>> I have even tried exactly Mark's suggestion (xtc file containing single
> >>> peptide) for test reasons, but a similar problem occurs for the peptide
> >>> denoted as chain N: do_dssp claims that are 124 residues in my selected
> >>> group, although 124 is the number of *atoms* (not of residues) of this
> >>> peptide. After the notion that some temporary file is backed up,
> >>> do_dssp gets stuck again.
> >
> >I saw a similar description in the archives a long time ago when I was
> > looking for help on do_dssp and experiencing similar problems (with a
> > simpler system, granted).  Which group are you selecting for running
> > do_dssp?  These errors seem to come up if you select any group other than
> > 'MainChain' (choice 5, I believe).  If do_dssp is finding an incorrect
> > number of "residues," perhaps you have selected 'Protein' instead of
> > 'MainChain' for your analysis?
>
> Thanks for your reply, Justin. Indeed I had chosen 'Protein' so far. But
> it is not evident that this caused the problem, since it worked well in
> some cases. Indeed, choosing 'MainChain' (indeed index group 5) does not
> help, but results in the same behavior.
>
> Volker


Greetings,


Florian

-- 
-------------------------------------------------------------------------------
 Florian Haberl                        
 Computer-Chemie-Centrum   
 Universitaet Erlangen/ Nuernberg
 Naegelsbachstr 25
 D-91052 Erlangen
 Telephone:  	+49(0) − 9131 − 85 26581
 Mailto: florian.haberl AT chemie.uni-erlangen.de
-------------------------------------------------------------------------------