In Lesson 1: 2D processing, display, and analysis, you will learn to process, display, and analyze 2D matrix files. The topics covered in this lesson are:
In Lesson 2: Analyzing Relaxation Data, you will learn to perform a relaxation analysis. The topics covered in this lesson are:
In this lesson you process a 2D matrix, display the spectrum using a variety of standard plotting methods, and assign the cross peaks. You start by processing a 2D matrix using the EZ macros. This lesson takes approximately 60 minutes to complete once you are comfortable with the procedures.
1. Setting up for the lesson
This lesson uses the 2D_tut tutorial data set. If you installed the gifts and tutorials, then you can find this data set in the $BIOSYM/tutorial/felix directory. If you did not install gifts and tutorials, then you can find this data set on the gifts and tutorials CD-ROM. Please copy these files to your working directory:
> mkdir pdata/1/
2. Starting FELIX
It is easiest if you move the parent text window that FELIX was started from to a position below the FELIX window. As you work with FELIX, information will occasionally be output to this window. Thus you should make sure that you can see this diagnostic information.
3. Reading in the first FID
This file is the first FID of the 2D HSQC spectrum collected on a Bruker spectrometer.
4. Apodizing the FID
Select the Process1D/Window Function menu item. Select Sinebell^2 as the apodization function. In the next control panel, leave the default parameters (512 and 90.0) and select the Real-Time option for Method.
The FID is displayed along with the apodization function in red. You
may experiment with different settings of the wsize and wshift parameters.
The effect on the FID is displayed in real time.
The apodized FID is now displayed on the screen.
5. Transforming the FID
6. Phase-correcting the spectrum
Select the Process1D/Phase Correction menu item. In the control panel, select the Real-Time option for Method and select OK. When the real time phase interface appears, <Shift>-click (with the right mouse button) a peak in the small overview window, which you would like to use for zero-order phasing. Using the mouse, adjust the Phase0 parameter to phase this peak, then adjust the Phase1 parameter as necessary. Select OK when you are finished.
Now that you have a rough idea of the apodization and phase correction parameters, you can proceed with processing the D1(t2) dimension of the 2D data set.
7. Processing the D1 (t2) dimension of the 2D data set
Select the ProcessND/Open and Process 2D menu item. In the first control panel, select the previous ser file. In the next control panel, leave the header parameters at their default values (read from the spectrometer header files), except for these:
There is a slight delay as the bruker.mat matrix is built. The bruker.mat matrix is then opened
As the D1 transform proceeds, the current row numbers are shown in the text window. This step should take about 1-2 seconds.
After the first dimension is processed, FELIX shows the control panel for processing the second dimension.
8. Processing the D2 (t1) dimension
The column numbers of the D2 vectors are shown in the text window as the processing proceeds. This step should take less than one minute.
When processing completes, the matrix is closed. If you choose instead to display the matrix at completion, the matrix is left open, the contour threshold is calculated, and the matrix is then displayed.
9. Opening the 2D matrix
10. Display the 2D matrix
FELIX automatically calculates the plot levels. You can change them manually:
The full 2D spectrum appears.
The data are always read from the matrix, not from a graphics file, so that display parameters can be recalculated at any time. Hence, the graph is redrawn with each plot command.
At this point you could examine the D1 and D2 vectors in more detail to determine if further baseline correction or phasing adjustments are necessary.
11. Referencing the matrix
Referencing of the matrix happens automatically, since the header parameters are read or adjusted during processing. You can further adjust the referencing, for example, by giving more descriptive names for the axis:
The 2D spectrum should now be redisplayed with the correct referencing for each axis.
12. Viewing an expanded-region contour plot
You can choose expanded regions with the cursor or by inputting numeric parameters:
When you release the mouse button, the region selected expands to fill
the window.The plot is still in intensity mode, hence the contour levels
are not shown.
The parameters in the control panel that appears are filled in with the values of the current plot.
The new 2D region appears.
You can save these parameters and reuse them for other plots, for example, if you were analyzing a series of spectra collected with different mixing times and always wanted to observe identical regions.
The 2D matrix is now plotted in contour mode, with a color-coded intensity scale.
13. Changing the 2D drawing parameters
If no peaks are visible, try decreasing the contour level to cut lower into the spectrum. If the peaks are outlined but you do not see the circles shrinking to define the tops of the resonances, try increasing the level multiplier to increase the space between levels:
14. Showing the grid display
Tick marks are now displayed
15. Plotting a stacked plot
The selected region of the 2D spectrum is displayed as a stack plot.
16. Returning to the full spectrum
Note: If you select Full Limits while in contour mode, the spectrum can take a long time to redraw. You can stop plotting by pressing <Ctrl>-c on your keyboard.
17. Picking peaks
Red boxes appear around all cross peaks meeting the criteria defined in the control panel. After the peak picking is done, a spreadsheet appears, displaying all the peaks.
18. Deleting peaks and replacing them
Select the Peaks/Edit menu item to manually adjust the box defining a peak. Click in the center of any red box of a picked peak. The box becomes green. Click the center of the box and drag to move the entire box or click near a corner of the box and drag to resize it.
19. Listing selected peaks
All the selected peaks from the peak-picking are colored with green boxes.
A summary of the current peaks in this list appears in a spreadsheet that you can edit.
20. Editing the list
There are now two or three fewer peaks in the list, corresponding to the peaks you have just deselected.
21. Assigning the cross peaks
You do not need to assign the peak in both dimensions. You can label the cross peaks in one dimension at a time, as the assignment is made. This is usually how assignments are observed.
If you want to use the restraints list directly in the Insight II or Discover program, you have to use the Insight II proton names as the peak names. At the moment there is no check of atom names, so you may enter anything you want.
22. Displaying assignments
The assignment list is searched, and the boxes surrounding all peaks with a label of h1 in D1 dimension are colored yellow.
23. Calculating volumes
To calculate the volumes of the picked peaks, select the Measure/Integral/Volume menu item. In the control panel, select the Measure All Volume options for Action. In the next control panel, leave the Peak and Volume set at their defaults. Set Volume Slot Number to 1 and Mixing Time to 0.1. Select OK.
To calculate restraints from these volumes based on the two-spin approximation, you must open or create a scalar entity for the database, define a scalar pair, create the restraints (strong-medium-weak, or any other listed choice), and write the restraints file. The appropriate commands are in the Measure menu.
24. Quit FELIX
In this lesson you learn to perform a relaxation analysis based on analyzing heteronuclear relaxation data. It is assumed that R1, R2, and heteronuclear NOE were measured as a series of 2D HSQC (or equivalent) spectra. The data used are parts of the relevant spectra acquired for apocalbindin D9k (Akke et al. 1993).
1. Setting up
2. Start FELIX
3. Access the Relaxation menu
4. Measure peak heights in the R1 series of spectra
FELIX now plots the spectrum, repositions the peaks to their exact centers. and calculates the peak heights.
When the spectrum selection control panel appears again, you need to specify file names and parameters for the remaining spectra in the same way as you did above.
You will find the relaxation delays in the last row of the table.
5. Evaluate the signal/noise ratio for the peak heights
FELIX calculates the peak heights in this duplicate spectrum, calculates the average height difference between this spectrum and its twin spectrum, and derives the uncertainty of the volume determination.
These values are reported in the text window. If you have more than one duplicate time point in your relaxation series, the uncertainties for the other time points are interpolated or extrapolated. For a single duplicate measurement, the uncertainties are simply promoted to the other points.
You will find the uncertainties in the last row of the table.
6. View a time course
FELIX plots a graph of the peak height vs. relaxation delay, including error bars. Due to the good S/N in the spectra, the error bars may not be immediately apparent.
FELIX displays the corresponding time course or informs you that no such peak number exists and lets you try again.
FELIX uses data from the currently active tables to display the time courses. If you want to view data from different tables, use the Preference/Table... menu item to assign another relaxation table.
7. Fit R1 values to the time courses
FELIX now fits the time course data to the exponential function:
and derives the relaxation rate R1 from the coefficient a2 in the exponent. The relaxation rate R1, its standard deviation, and the c2 value for each fit are reported in the text window.
The relaxation rate R1 and the raw coefficients for the offset a0 and linear term a1 in the function, along with their standard deviations are stored in the table rel:r1. In addition, the c2 value of the fit is stored for each time course.
If you now view a time course, FELIX plots the fitted function in red, along with the experimental peak heights. This lets you visually judge the quality of the fit. You can print the plot by clicking the printer icon.
8. Evaluate R2 data
The t2_1b.mat is a duplicate spectrum. R2 time courses are fitted to the simple exponential function:
and the general exponential function:
Whichever function yields the lower c2 value is used to derive the R2 relaxation rate. FELIX reports the R2 value, its standard deviation, and the c2 value in the text window and also tells you which function was used.
All the fitted values are stored in the table rel:r2, analogous to the R1 data (see Step 7).
9. Evaluate heteronuclear NOE's
FELIX plots the first spectrum, repositions peaks to their exact centers,
and measures peak heights. Then it plots the second spectrum and measures
FELIX plots the first duplicate spectrum, repositions peaks to their exact centers, and measures peak heights. Then it plots the second spectrum and measures peak heights. Finally it reports the NOE's and their standard deviations to the text window and stores them in the table rel:noe.
10. Generate Modelfree input
Once you have all R1, R2, and NOE values evaluated and stored in the
database, you can generate an input file for the Modelfree program (A.
G. Palmer, Columbia University, http://www.hhmi.columbia.edu/
Now you have initial input files for the tmest and Modelfree programs. For more information about working with Modelfree, please refer to its documentation and to the scientific literature.
11. Exit FELIX