Hello, Tim says (if any questions still then email again):

Usually asssigning H & N atom types to the HSQC is the first thing that is
done (Initialise Root Spectra before sequential assignment), so subsequent
linking is not normally needed. This system also takes care of amide side
chains.

Otherwise, setting of atom types quickly for H & N can be achieved by
setting the atom types in the resonance table, after selecting only the
aprropriate rows.

At present, propagation of specific atom assignments only presently occurs
through direct 'onebond' links, i.e. not TOCSY.

Other transfer types are simply not considered. The potential presence of
NH2 side chain groups in HSQC spectra complicates the code for the
automatic setting trough TOCSY transfers. And anything automatic must be
safe. Other transfer types could be considered, but I am reluctant to
consider this a high priority for amides which can be set early on, or en
masse.

The warning message is due to checking chemical shift values against the
BMRB distributions. This check happens each time you assign an atom type.
The warning is triggered by the load of the reference information for each
residue type, but is otherwise unconnected to the actual assigment
process.

Wayne

On Fri, 6 Jun 2008, Eliza Ploskon wrote:

> In my project file I have few simple spectra: hsqc, cbcaconh, hncacb and
> hcch-tocsy and all experiments have their types specified. I have made seq
> linking, predicted type of spin systems and assigned sequence. As for me
> that should be enough for the program to know that it is dealing with N, HN,
> Ca, Cb types of atoms since it knows what experiment it is. When I assign Ha
> or Hb protons in tocsy analysis only assigns carbon resonances but not amid
> nitrogen and proton. Why?
>
> I also got a message:
>
> Deprecated: getByKey called with too short key
> from doLoad line 770 in
> /home/chzeap/ccpnmr/temporaryReleaseDir/ccpnmr/ccpnmr1.0/python/memops/format/xml/XmlIO.py:
> None
>
> I know form previous emails that this "error message is a harmless warning"
> and "should have nothing to do with the behaviour i m observing" but i do
> not quite understand how it has nothing to do with my problem, since it
> appears every time when analysis should assign amid N and H but it doesn't
> do it.
>
> many thanks for any suggestion how to fix it, so i do not have to do it manually
>
> Eliza
>