For CEST on amide groups in proteins, data should be collected using the experiment from
Vallurupalli, Bouvignies, and Kay, JACS 134, 8148-8161 (2012)
Our version of this pulse sequence is called CEST_15N_protein_1.rp
The CEST experiments should be processed with nmrPipe. An initial processing script can be generated from within Topspin3 using the au program convert2pipe.ecj33. You can use standard processing initially.
Note: all of the files and scripts discussed below will be placed in and run in the Main NMR Directory. This is the Topspin directory that contains all the data from the series you’re analyzing (ie the <expt name> directory).
nmrPipe has tools to analyze 2D relaxation data. You can find a description of it on the nmrPipe website. Follow the instructions for using the GM window function during nmrPipe processing. Then use proc.com, fid.com, and ft2.com to process all spectra in the series. tauList and dirList inside proc.com have to be modified to include your specific CEST offsets and the corresponding directories. If you ran one experiment with no CEST period, then make the first offset “-10000”, as a marker, then all the other offsets after that one. Run by typing ./proc.com. After that, pick and assign peaks in one of the spectra (using nmrPipe). nmrPipe calls the saved peak list “test.tab” by default. Rename this file relax.master.tab, and place it in the Main NMR Directory. Then run fit.com (by typing ./fit.com). It can take awhile to run (sometimes overnight for CEST data). Just be patient and wait until the little blue window disappears and you get the prompt back. It will generate two new files called axt.tab and nlin.tab.
I have written Python scripts to take all the data from nlin.tab (or axt.tab), and order it appropriately for further analysis. Place the script order_CEST_intensities_nmrpipe.py in the Main NMR Directory. Inside the script, modify the line that starts offsets = range(). For example, if your offsets were -1000, -975, -950… 975, 1000, then that line would read offsets = range(-1000, 1000, 25). Just after that, there’s a line that reads: offsets.insert(0, -10000). If you ran one experiment with no CEST period, then you should uncomment that line so that it inserts the offset value -10000 at the beginning of the list. This offset value is a flag in ChemEx that indicates the experiment run with no CEST period. Create a directory called output_files, then type ./order_CEST_intensities_nmrpipe.py to run it. This will create a separate file for each residue (in the directory output_files). These files have 3 columns: 1) The CEST offset, 2) The intensity, 3) An error estimate.
Run the script graph_intensities.py to see graphs of the peak intensities CEST offset for each residue. These scripts generate the graph for one residue at a time. When you click the graph for the 1st residue away, the next one will appear, and so one. This is just a rough way to visualize the data. If some residues look promising, you can proceed with analysis using ChemEx or other software.
Example scripts are in the directory /home/peterson/NMR/CEST_analysis
Scripts:
fid_v33.com (initial nmrPipe processing script)
proc.com (proc.com is a script that contains the CPMG “frequencies” and a list)
fid.com (of the directories containing the experiments done using those frequencies)
ft2.com (fid.com and ft2.com are nmrPipe processing scripts that use proc.com as input)
fit.com (script that fits intensities from all spectra in the group to a decaying exponential)
order_CEST_intensities_nmrpipe.py (Python script that orders the intensities in a separate file for each residue)
graph_intensities.py (Python script that graphs the intensity vs. CEST offset for each residue separately)
