backbone assignment doesn't find matches

I’m trying to assign a backbone but have discovered that analysis is not always showing good matches. I’m matching CO, CA and CB shifts, but I’m only shown poor matches. Attaching a figure showing the CO region only (blue = HNCO, red = HNCACO) but CA and CB matches are also terrible.

However, if I set only to match CO shifts, I get good matches - both in the CO region, but these residues also match perfectly for CA and CB also (now shown):

What’s going on?! I’m halfway through an assignment and now I’ve discovered this I have to go back to square one because I can’t trust I haven’t missed alternative matches!

(Also - is there no way to show CO and CA/CB in separate windows so I don’t need to constantly scroll up and down to check matches?)

Following on from this… I can’t even figure out how to remove all my assignments so I can start again. If I select deassign it deletes everything - atom assignments etc. What should I do?

Thanks,

Chris

Hi Chris,

Following on from this… I can’t even figure out how to remove all my assignments so I can start again. If I select deassign it deletes everything - atom assignments etc. What should I do?

On this front: What you probably want to do is go to the Sequence Graph and right-click on an NmrResidue and select Disconnect all NmrResidues (or obviously you can just break selected links if you want).

Screenshot 2024-11-07 at 11.18.53

Vicky

On the other issues:

As @bosmith mentioned in his e-mail, one thing that will help you, is to add a Search SpectrumDisplay.
I think we added the “Show Search Strip in Match Module” because someone asked for this, but probably didn’t really think through that people would then not also want to have a separate Search SpectrumDisplay (though that does obviously make sense!). I’ll look into temporary and longer term fixes for this.
For the time being, I would suggest that you open an another 3D SpectrumDisplay and set this as your Search SpectrumDisplay. If you want, you can detach it from the Drop Area (right click on the blue/purple bar at the top and select Detach from Drop Area) and then just leave it somewhere in the background so it doesn’t take up screen space. Having this present (even if only in the background) will ensure you have all the matching lines drawn in correctly etc. which will make life much easier and hopefully also sort out the dodgy matching.

As regards separate windows:
Yes, you can do this. Synchronising Axes (see https://www.ccpn.ac.uk/manual/v3/SynchronisingAxes.html) will allow you to have two windows which go to the same H and N (X and Z) positions, but different C (Y-axis) positions.

Let me know how you get on with this and if you need any further help.

It looks to me like there is also a bug when you select not to focus the Y-axis (it possibly seems to go to a default value?), so we’ll look into that, too.

Vicky

Hi @bosmith @VickyH,

Thanks for the responses - copying onto here from our email thread. I still have the same fundamental problem - analysis only correctly finds matches when I exclude both CA and CB atoms from the search. The user interface is definitely problematic, but a secondary issue!

Possibly related - the sequence graph does not show any links involving CB…

PastedGraphic-1

Thanks,

Chris

Just in case anyone encounters the same problem, the solution was to make sure that the Reference Experiment Dimensions are set for the experiments that you’re trying to assign with.

Sounds like we need some better error messages?