Saturday, May 22, 2004
Correlation between TOC and NOE Data
One of the problems of the two dimensional Nuclear Magnetic Resonance (NMR) data that I am looking at this morning is that a significant number of peak contours are convoluted between multiple peaks. I have been able to use the correlation of peak data between Nuclear Overhauser Effect (NOE) and TOtal Correlation SpectroscopY (TOCSY) contours to discriminate between multiple peaks. The key is that the TOC data does not include peaks that relate the N and N-1 residues that arise as a result of nuclear relaxation through dipole-dipole coupling of those neighboring spin systems (Neuhaus and Williamson, 1989). The TOC data are only the result of though bond scalar coupling. That means that I can use the TOC data to pick the within residue peak, in the NOE dataset, when it and a neighboring linkage peak are nearly on top of each other.
A practical problem arise however, because this observation is implemented directly with human observation and human peak additions that I can confirm in NMRView. I am not sure how to automate this part of the process. I would require either faith that I could write a bullet proof rule or do some peak shape analysis. Both solutions look like the process could be spoofed. For now I will just do the peak picking by hand.
Add your comment
(0) comments
One of the problems of the two dimensional Nuclear Magnetic Resonance (NMR) data that I am looking at this morning is that a significant number of peak contours are convoluted between multiple peaks. I have been able to use the correlation of peak data between Nuclear Overhauser Effect (NOE) and TOtal Correlation SpectroscopY (TOCSY) contours to discriminate between multiple peaks. The key is that the TOC data does not include peaks that relate the N and N-1 residues that arise as a result of nuclear relaxation through dipole-dipole coupling of those neighboring spin systems (Neuhaus and Williamson, 1989). The TOC data are only the result of though bond scalar coupling. That means that I can use the TOC data to pick the within residue peak, in the NOE dataset, when it and a neighboring linkage peak are nearly on top of each other.
A practical problem arise however, because this observation is implemented directly with human observation and human peak additions that I can confirm in NMRView. I am not sure how to automate this part of the process. I would require either faith that I could write a bullet proof rule or do some peak shape analysis. Both solutions look like the process could be spoofed. For now I will just do the peak picking by hand.
