Processing a 3D CBCA(CO)NH experiment
The sample was 13C,15N-ubiquitin run on a Bruker 600. (Data kindly supplied by DuPont Pharmaceuticals.)
The data set is 1H-detected (512 complex points), with 13C in the second dimension (110 total slices, hypercomplex) and 15N in the third dimension (96 total slices, hypercomplex) for a total size of 43 MBytes. It was acquired with digital filtering, so needs to be processed with RD (see details). There is also a residual water signal which needs to be removed to avoid some truncation artifacts, so we use a digital high-pass filter (DH). (Here is an illustration of the truncation artifacts that can result.)
See macro for complete processing with a single command, but this can also be done from the keyboard, as described below.
Using a macro, processing took 42
seconds on a Pentium II/400, including RD and DH operations. ![]()
The data was imported with IM, which finishes by displaying the first slice.

Processing parameters are determined by processing the first slice. The parameters used were cosine multiplication (set S# to 90, them MS), RD to correct for artifacts of the digital filter, DH (filter width of 250 Hz) to remove the residual water signal, FT and phase. It is important to examine the FIDs and tailor processing to avoid artifacts such as truncation. (See example.)

The data was nominally a 2D data set, with 512 complex points and 10,560 slices (110 x 96) in a Bruker ser file. Nuts imported this as a 2D file, so the dimension sizes must be set correctly. To do this, enter non-2-letter command mode by typing 2F, and then use the dims command.
dims <enter>
will print out the number of points in each dimension, which for this data set is
dims - Points 1D = 512 2D = 10560 3D = 1 4D = 1
To set these correctly, use dims and enter the 4 values, all on one line, separated by spaces:
dims 512 110 96 1
Once a value greater than 1 is entered for dimension 3, the View/Spectral Parameters menu will display parameters for all 3 dimensions:

Note that the 3rd dimension did not contain correct values for SF and SW. This is because the raw data did not include any 3D parameters. The correct values were entered using this screen.
We are now ready to process. The first step is to enter "3D arrayed mode", by typing
ar 3d <enter>
The following series of commands is executed. While not specified, each command is terminated by <enter>. See Nuts Help for explanation of each command. (Remember that the data is hypercomplex in both indirect dimensions.)
ms
rd
dh
ft
ps
tr
tr
st
Now we are done with the proton dimension, and need to transpose the data to process the carbon dimension. Because we will need to "build" complex interferograms in the nitrogen dimension (using tr tr st commands), we need the carbon dimension to be dimension 1 and the nitrogen dimension to be dimension 2. This is accomplished by 2 successive TD commands. The TD command, when working with 3D data, takes an argument to specify which dimension will be transposed with the current dimension 1.
td 3
This puts the proton as dimension 3 and nitrogen as dimension 1.
td 2
This puts the carbon as dimension 1 and nitrogen as dimension 2. Now process with
ms
ft
tr
tr
st
This data was collected to require no phasing in the carbon and nitrogen dimensions. We have constructed our complex interferograms in the nitrogen dimension, so we transpose to process the nitrogen dimension.
td 2
And process with
ms
ft
And a final TD to put the nitrogen back to dimension 3 and display protons as dimension 1:
td 3
Now we are ready to view the data. IP will display the intensity plot, and the commands
plane +
plane -
will step through 15N planes. The number of the currently displayed plane is shown in the upper left corner of the screen.

The zoom subroutine works just as for 2D data to expand the display.

Exit from the IP routine and use SP to view stacked plots. The ] and [ keys will step through planes.

Last updated: 01/22/03
Acorn
NMR Inc.
7670 Las Positas Rd
Livermore, CA 94551
(925)456-1020 FAX
(925)456-1024