Processing 2D data
The user must have a basic understanding of how the data was collected to select the correct processing method. For example, just knowing that the experiment is COSY is not enough, because the data may or may not be phase-sensitive. The data types that can be processed with NUTS are magnitude, hypercomplex (method of States, et al), echo-antiecho, States-TPPI and TPPI. Each type must be processed differently.
Sample macros with detailed explanation is provided for those experiments for which we have sample data. We will create macros for the others if sample files can be supplied. Contact us for help in creating customized macros.
A recommended reference is T.D.W.Claridge, "High-Resolution NMR Techniques in Organic Chemistry", 1999.
Step-by-step instructions for:
Magnitude data - COSY or HETCOR
Hypercomplex data
TPPI data
Echo-antiecho data
States-TPPI data
** NUTS-Pro versions newer than May, 2002, include a modified arrayed mode for improved processing of phase-sensitive data. See details.
See also:
arrayed mode processing (NUTS-Pro users only)
macros - commands used in macros for automated processing
2D processing - commands related to 2D
displaying 2D data - options for contour plots
phasing 2D data
symmetrizing
editing 2D data
sample data for downloading
Varian DEPT data - macros for processing and creating edited DEPT plots
comparing multiple 2D spectra
Varian 2D data
Processing description and macros are provided for the following Varian experiments. The same procedures are applicable to data from other spectrometers.
NOESY - nuclear Overhauser spectroscopy; hypercomplex, phase-sensitive using States method
The same processing procedure applies to tntocsy, tocsy, roesy, and hypercomplex cosy data.g_hsqc - gradient heteronuclear single-quantum correlation; echo-antiecho, phase-sensitive
hsqc - heteronuclear single-quantum correlation; hypercomplex, phase-sensitive
hmbc - heteronuclear multiple bond correlation; hypercomplex, not phase-sensitive
g_hmbc - gradient heteronuclear multiple bond correlation; magnitude
cosy - magnitude COSY, not phase-sensitive
2dexch - 2D exchange experiment for solids; hypercomplex data in which the 2 halves of the data are scaled differently
magnitude -- no phasing, a magnitude calculation is done at the end of processing; This is appropriate for experiments such as cosy, cosygs, cosygsmf, cosy45gs, inv4gs, invgslrlp. See also hetcor processing.
phase-sensitive experiments:
TPPI -- requires REAL ft in the indirect dimension
States (hypercomplex) -- pairs of slices are combined to generate complex "fids" in the indirect dimension
echo-antiecho -- gradient experiment, pairs of slices are added and subtracted to generate complex "fids" in the indirect dimension
states-tppi data -- experiments such as noesygpst and roesyprst
For Avance-series spectrometers, you can tell which kind of data you have by opening the file called pulseprogram using a text editor such as Word or WordPad (notepad doesn't work well, due to the absence of carriage returns). Toward the end of this file is a parameter called MC2, which indicates the type of processing required.
MC2 values:
QF indicates magnitude data
QSEQ or TPPI indicates TPPI data
States indicates States-type hypercomplex data
States-TPPI indicates States-TPPI data
Echo-antiecho indicates echo-antiecho data
The name of the pulse program also indicates the data type. The names are concatenation of a base name (usually 4 letters) plus a series of 2-letter codes that are shorthand notation for various properties of the experiment. Codes of interest here are:
ea -- echo-antiecho
gs -- gradient selection (use magnitude processing)
sh -- use States-type processing
tp -- use TPPI processing
st -- use states-tppi processing
For Aspect-based spectrometers, data is most likely magnitude or TPPI, depending on whether or not it is phase-sensitive. Reading the .aur file on the spectrometer will usually tell you which. Or, process the first slice of the data and try to phase it; if it can't be phased, assume it's magnitude. If it can be phased, assume it's TPPI.
How to tell if something is wrong
If the wrong type of processing is done, a set of mirror image peaks is often seen. Compare the plots shown below for an HMQC spectrum of strychnine. The top spectrum is correct. The bottom spectrum was obtained by failing to process the data as hypercomplex.
Because there are so many variations in spectrometer models and pulse sequences, it is not possible to provide users with macros that will work correctly on 2D data from all spectrometers. It is recommended that users process a known data set first, for each 2D experiment commonly used on their spectrometers, to determine the correct processing commands under conditions where the correct final result is known.
Another, less serious, problem is that sometimes the final spectrum appears to have the diagonal going the wrong way, as shown below.
This is easily fixed by including a spectrum reverse (SR) in the processing for the second (indirect) dimension. Or, if there is already a SR in the processing list, remove it and re-process.
If the data is homonuclear and "square" (same number of data points in both dimensions), the spectrum can be symmetrized using the SY command. Choose Spectral Parameters from the View menu and check the Number of Points in the 2 dimensions. Additional zero-filling can be done during processing to make the data end up square, if symmetrizing is desired. Attempting to execute SY if the data set is not square will generate an error message.
Last updated: 12/4/02