co-MDD processing of a pseudo-3D experiment

[Guillaume Bouvignies]

Hello Maxim,

I have recorded a relaxation experiment (15N CEST, Vallurupalli et al., JACS, 2012) using the NUS functionality of Topspin. The result is a pseudo-3D experiment (one "ser" file) with interleaved planes (1H-15N hsqc-like) where peaks share the same shape and position, but where their intensity differs from plane to plane. As the points were sampled randomly in both the indirect and the pseudo dimensions, each plane ends up having a different number of points.

I believe this is a similar problem to the one you dealt with in your recent paper about time-resolved NMR spectroscopy (JBNMR, 2014 ; very nice work by the way!). I tried to use qNMR to process my data, but I couldn't find a way to tell the program that the pseudo-dimension is particular and, basically, only encodes an amplitude profile for each resonance (A_i in your paper). Would you have some template files showing how to run mddnmr to process this kind of experiments?

Thanks,

Guillaume

[Maxim Mayzel]

Dear Guillaume,

Indeed your problem sounds similar and you definitely would benefit from co-processing, although the spectrum you have is quite different. In TR spectrum all points are "in order", every consecutive data point in the spectrum corresponds to the next time interval so the signal intensity decays. In your spectrum, as far as I understood, consecutive data points are randomized and I don't have a script to put them in the order. Instead I would advice you to split pseudo-3D spectrum to individual planes (use nmrglue, if you are familiar with python) and then just do co-processing, and for this you can use qMDD interface. 

Would you need any other help or advice feel free to contact me or Vladislav directly, although I'll be away for approximately two weeks.

Best regards,

Maxim