Reply from Tim:
> Date: Tue, 2 Feb 2010 08:58:27 -0400
> From: David Langelaan <[log in to unmask]>
> Reply-To: CcpNmr software mailing list <[log in to unmask]>
> To: [log in to unmask]
> Subject: Bugs and possible features
>
> Possible Bugs:
>
> 1. In peak tab of the quality reports, every single peak from my HNCO
> spectrum is flagged with 'Impossible bond 57 LysC-58GlyN' etc... This
> experiment has been set to HNCO, so I doubt this is the problem. As far as I
> can tell, this is a perfectly normal connection and is being improperly
> flagged, although I may be wrong.
This is fixed in the updates: some old code was lurking which needed to be
removed, to make way for the proper way of detecting bonds across residues.
> 2. This problem is a bit more of an enigma. During my initial assignment I
> set some protons to be the HD1 proton of histidine. I then created distance
> restraint lists from the NOE data. Later I realized, that the HD1 proton was
> actually the HD2 proton, so I renamed the appropriate resonances, which were
> then correctly updated in the restraint lists. The problem is that when I
> export the restraint list, it is exported as the initial HD1 instead of HD2.
> Also, if I analyze violations a pdb file, the software looks for a HD1 atom
> instead of the HD2 atom which the restrain list shows.
As Rasmus mentions this is due to the Restraint Set having a fixed copy of the
initial assignments. In the latest updates I have adjusted
Structure::Restraints & Violations::Restraints:[Update Assignment From Peak],
so that it now not only refreshes according to the peak-resonance assignment,
but also the atom-resonance assignment. If you know of a specific atom
assignment change this is easier than using a new Restraint Set.
> 1. In the resonances tab of the quality reports, some resonances are flagged
> because one of their peaks have a large delta. However, when I click show
> peaks it can be tedious to find out which peak exactly is the problem. Would
> it be possible to include a setting to pull up the violating peak or at least
> highlight it?
The way to get at this info from the Quality Reports is to use the [Show
Resonances] button, then in the resonance table that appears click the row you
are interested in and then [Info] - this will show all the positions of the
resonance's peak dim assignments so you can easily spot the wayward ones.
> Some of my resonances are flagged in the quality report because they have
> multiple bound partners. Say a glycine Ha bound to both a Ca and a Cb. I
> understand how this is a problem, but I have not found a way to fix it. I am
> not sure where this error is coming from, since in some cases there are not
> any peaks which connect a resonance to the improper bound partner.
As Rasmus may have mentioned this sort of thing has two common causes:
Genuine assignment mistakes (the main point) - you'll have to look
though the peaks to fix these
Problems with the reference experiment setttings: usually because there
are two dimensions with the same isotope that are swapped compared to
what is described in the prototype.
T.
|