Hello,
I wondered what’s the easiest way to transfer the assignment of a 3D noesy-hsqc to another 3D noesy-hsqc of the same protein, collected for example in a slightly different buffer composition and/or pH. In this case I have mainly the same peaks but with small shifts.
Shall I calibrate the noesys so to have the hsqc-like projection be matching?
Many thanks!
Francesca
Hi,
As usual the answer is…it depends!
The main considerations are how different the chemical shifts are and what you’re going to do with the information from the NOESY.
If you have NOESY data I’ll assume that you are heading towards structure calculation.
If the shifts are really small so that they’re within the tolerances of your peak matching algorithm in your structure calculation protocol (often something like +/- 0.04 for 1H and +/- 0.3 ppm for heavy atoms) then you probably don’t need to do anything.
If some of your shift diffrences are larger than this and it’s a 15N edited experiment then assigning the matching 15N HSQC (or projection) will sort most of your assignments out (since H & N are typically much more affected by buffer conditions & temperature than aromatics and aliphatics are).
Best practice for managing spectra recorded under different conditions is to group them in different chemical shift lists (one list per condition). We’d recommend keeping your spectra correctly referenced rather than fiddling the referencing to get a good match. So the question is, how to get the non-shifting (aromatic and aliphatic) shifts into your new chemical shift list? You can temporarily move the spectra with the fuller assignment into your new list and then move them back to their correct shift list - the shifts of NmrAtoms that aren’t (yet) assigned on the new spectra will be populated in the chemical shift list.
If you’re using spectra curated this way to generate restraints for structure calculation you will need a workflow that accepts both the peak lists from the spectra and the chemical shift list as input.
I hope this helps.
Hi!
Many thanks for your answer.
I add some details of what I am doing.
In the past, have collected a set of data on my protein, assigned the 3d noesy (both 15N and 13C), and have used the restraints for structure calculations.
Now, for a new project, we are working on the same protein, but we switched to a different buffer composition and pH due to experimental reasons.
We then also collected the 3d noesy in the new conditions, because we might need it even to just confirm the peaks positions in the 2d trosy spectrum, if they are moving a lot.
Therefore, I would like to have the new 3d 15 noesy assigned.
I transferred the peaklist from one 3D to the other and I have a peaklist on the new noesy, but of course, due to the shifts of the peaks in the spectrum, the assignments are often out of plane and it is difficult to go back to the assignment for very close peaks.
We are now doing a lot of protein/ligand studies and in some cases, we also need the noesy to be collected, so it would be very useful to have a correctly assigned 3d noesy in the starting conditions.
Therefore I wondered what’s the quickest way to transfer the assignment in the 3d 15N noesy, to have the assignment correctly positioned in the strips, for all the residues.
Many thanks!
Francesca
Since you’re probably not going to recalculate the structure for the new conditions (yet) you are mainly using the NOESY for confirming the assignments under your new conditions, I would recommend:
transfer the assignments between the HSQCs (or HSQC projections) for the two conditions (see the beginning of the chemical shift perturbation tutorial at https://ccpn.ac.uk/support/tutorials/ for how to do this) for the easy to match peaks
for the more shifted/ambiguous HSQC peaks assign them to new placeholder NmrResidues and NmrAtoms (the @ types) and then use these to navigate into the 3D to find the crosspeaks that will help you to confidently assign them. Merge the placeholders with the confirmed assignments once you are confident
Only peak pick and assign the whole 3D if you are going to actually use that data.
Thanks a lot!
This seems to be the most reasonable procedure, indeed!
Many thanks!
regards,
Francesca