No part of this document may be reproduced, transmitted, transcribed, stored in a retrieval system, or translated into any language in any form by any means without the written permission of Spectrum Research, LLC.

Spectrum Research, LLC. reserves the right to change the information in this document without prior notice.

Trademarks

SpecMan and NMR-SAMS are trademarks of Spectrum Research, LLC.

Acknowledgments

Computer-Assisted Structure Elucidation of Q-2
Using SpecMan and NMR-SAMS

Overview:

The following document provides step-by-step instructions that lead the user through the process of computer-assisted structure elucidation of the betulinic acid (Q-2) molecule. We assume the users of this document to have a good understanding of general NMR techniques and their application to structure elucidation problems. There are two parts to this document: Part I describes step-by-step instructions for each task in SpecMan, while Parts II describes step-by-step instructions for each task in NMR-SAMS.

Q-2 (Fig. 1) is a natural triterpene isolated from American white-barked birches (Betula papyrifera Marsh., Betulaceae) and was provided by the group of Dr. N. Farnsworth at the University of Chicago. It exhibits a molecular formula of C₃₀H₄₈O₃ (MW = 456), based on ¹³C-NMR and DEPT spectroscopy and EI-MS. Its identity with betulinic acid was proposed by comparing its physical data as well as NMR spectral data with those reported in the literature (S. Siddiqui et al., J. Nat. Prod. 51, 229 (1988); M. Sholichin et al., Chem. Pharm. Bull. 28, 1006 (1980).) Because of the severe overlap in resonances, there had been no report of complete ¹H and ¹³C assignments for this compound until the recent studies of the molecule at higher magnetic field (C. Peng et al, Magn. Reson. in Chem., 36, 267-278 (1998).).

Although this tutorial is organized in such a way that peak picking using SpecMan is described in Part I, and structure elucidation using NMR-SAMS is described in Part II, it is highly recommended that the user run both programs side-by-side.

Figure 1. Two-dimensional structure of Q-2 with the ¹³C and ¹H (in parenthesis) resonance assignments. The numberings of the atoms corresponds to those of the ¹³C and ¹H peaks in Tables I and II.

Part I: NMR Analysis and Assignment with SpecMan

This part provides step-by-step instructions for computer-assisted peak picking with SpecMan.

I-1. Transferring Processed Spectra from NMR Spectrometers

The spectral data of Q-2 was acquired on a Varian Unity-plus 720 MHz spectrometer and processed (FT, phase correction, etc.) with VNMR 5.1. The concentration of the sample used was 0.084 M in pyridine-d₅. In order to import Varian processed spectra into SpecMan, the user needs to transfer the phasefile and the procpar files for each spectrum to the working directory on the PC.

For this tutorial, the sample data is located in the Data directory on the CD. Copy the Data.zip file from the Data directory into the Spectrum2001 directory on your PC. Then extract the zip file using an extraction program, such as WinZip. The default location for the sample data is: C:\Spectrum2001\Data\SpecMan\Q-2 for SpecMan data, and C:\Spectrum2001\Data\NMR-SAMS\Q-2 for NMR-SAMS data. The following subdirectories will appear under the C:\Spectrum2001\Data\SpecMan\Q-2 folder:

H-1

C-13

DEPT

COSY

HMQC

HMBC

NOESY

I-2. Analysis of 1D Proton Spectrum

From the Program Manager or the Start Menu, click the SpecMan icon in the Spectrum2001 group to launch the SpecMan program. Then, open a 1D ¹H file by selecting ‘Open Spectrum’ from the File menu. Select the File Type as Varian, and then use the file browser to change to the H-1 directory and double click on the phasefile in this directory as shown below:

After double clicking on 'phasefile', the 1D spectrum will be displayed in a 1D Slice Window as shown below:

Setting Reference

The next step in analyzing the ¹H spectrum is to set the reference. The original reference parameters are obtained from the procpar file, but the user may want to change the spectral reference.

In this case, zoom on the small, weak peak due to residual g-proton of C₅D₅Nat about 7.55 ppm by using the right mouse button to select the left top corner and drag the mouse (keeping the right mouse button pressed) to the desired right bottom corner (a rectangular zoom box will be drawn) and release the button. The selected peak will be expanded and redrawn in the window. Next, select ‘Set Reference’ from the Edit menu. Place the cursor at the top of the peak and click the left button. The following ‘Set Reference’ dialog box will appear with the X chemical shift of the peak:

If the value is not already 7.55, then type ‘7.55’ in the X Reference ppm text box, and click, ‘OK’.

To reset the expansion to full view, select Reset Zoom in the Display Menu (or click on the ‘Reset Zoom’ icon on the tool bar). Once the reference has been set, the relevant spectral parameters (including reference) will be saved in the procpar file.

In this case, peak picking of ¹H NMR spectra is futile due to severe overlap in the ¹H peaks. Therefore, it is preferable to use 2D HMQC to extract the 1D proton chemical shifts (see section I-5).

I-3. Analysis of 1D Carbon Spectrum

Setting Reference and Appropriate Threshold

To begin analyzing the 1D ¹³C spectrum, open the ‘phasefile’ from the C-13 directory and the following ¹³C spectrum will appear:

Similar to how the reference was set for the ¹H spectrum, set the reference on the middle peak of pyridine at 135.5 ppm. Next, set an appropriate threshold by selecting the ‘Set Threshold’ button from the 1D Control Panel, as shown below:

This will display a red horizontal line on the ¹³C spectrum when the cursor is moved to the spectral window. Move this line to a position just above the noise peaks and then left-mouse click to update the threshold value. One can also type the necessary threshold value in the threshold textbox on the 1D Control Panel (use 9.906e-003 for this example).

Automatic Peak Picking

To perform peak picking, select, ‘Pick Peaks Automatically’ from the Analysis menu with the pull-right option of ‘1D.’ The following dialog box will appear:

Uncheck ‘Negative Peaks’ and then click ‘OK.’ Thirty-eight peaks will be picked and displayed on the spectrum and listed in the 1D Peaks Table as shown below:

Manually Removing and Adding Peaks

Next, delete the three solvent peaks (triplets between 150 to 123 ppm; a total of nine peaks, numbered 28 through 36) by selecting ‘Remove Peaks’ from the Analysis menu. Using the left mouse button, define a rectangular rubber-band box around the solvent peaks. As soon as the left mouse button is released, the solvent peak labels will no longer be displayed on the spectrum, and the solvent peak entries will be removed from the 1D Peaks Table. Next, re-click on the ‘Remove Peaks’ button from the Analysis menu to deactivate the option.

SpecMan also gives the user the ability to add peaks by selecting ‘Add Peaks Manually’ from the Analysis menu with the pull-right option of ‘Without Refine.’ Then, click the left mouse button at the desired position to add a new peak. Please note that there are not any new peaks added to the ¹³C spectrum (We assume that the user has not yet identified the overlapping peak at 16.44 ppm. This will be pointed out later by NMR-SAMS and the peak will be split into two peaks).

Sorting, Editing and Saving Peaks Table

Next select ‘Edit Table’ from the 1D peaks table, and the following ‘Edit Peaks Table’ dialog box will appear:

Select the following options:

‘Sort Table Entries’

‘Descending’

‘X Value (i.e. sort the table in the descending order of ¹³C Chemical shifts)’

‘Renumber ID’s’

‘Remove Redundant Peaks’

Then, click ‘OK’ to sort the table, and the 29 peaks will be sorted and renumbered in the descending order of their chemical shifts. Next click ‘Save Table’ from the 1D Peaks Table, and save the peaks table as ‘c13’ (The extension *.pks will be added automatically).

I-4. Analysis of DEPT Spectra

The DEPT experiment usually consists of DEPT-45, DEPT-90, and DEPT-135. When a Varian spectrometer is used, the DEPT experiment is stored in the form of arrayed 1D experiments. SpecMan automatically detects the multiple 1D spectra and displays a slider in the following 1D Control Panel:

In order to view each experiment type, the user can move the slider on the 1D Control Panel to position 1 for DEPT-45, to position 2 for DEPT-90, and to position 4 for DEPT-135, or the user can view all three experiments by selecting, ‘Open Multiple Spectra’ from the File menu. This will bring up the following ‘open multiple spectra’ dialog box:

To add the first DEPT experiment, click on ‘Add’ and the following dialog box will appear:

Next, select the DEPT ‘phasefile’ from the DEPT folder. To add additional spectra, re-select ‘Add’, and then select the same DEPT ‘phasefile’ two more times (this is necessary since the DEPT experiment is saved as an arrayed spectrum). As you select the files in the ‘Open’ file browser, the list of selected files will be displayed in the ‘Open Multiple Spectra’ dialog box shown below:

Once you have selected the desired spectra, click ‘OK’ in the Open Multiple Spectra dialog box. The following warning dialog box will appear:

Click ‘OK’ and the following multiple spectra will be displayed:

To display the DEPT-45, DEPT-90 and DEPT-135 spectra, set the top spectrum as active (select ‘Set Active Viewport’ from the Display menu and then click on the top spectrum). Since DEPT-45 is the first position on the slider, it is already at the right location. Next, set the second spectrum as active and move the slider to the second position to display DEPT-90. Then, set the third spectrum as active and move the slider to the fourth position to display DEPT-135.

In order to keep the three DEPT experiments separate, rename the spectral titles by choosing ‘Set Labels’ from the Edit menu and rename the spectra as ‘DEPT 45,’ ‘DEPT 90,’ and ‘DEPT 135’, respectively.

When comparing DEPT spectra it is convenient to tie the axes display. This is accomplished through the ‘Tie Views’ option in the Display menu. When this option is chosen, the following ‘Tie Manager’ dialog box is displayed:

Select, ‘Tie All 1D’ to tie the X axes of all open 1D spectra. Note that in the tied mode the cross-hair cursors of each spectrum are also tied together so that they move together when you scroll in a spectrum. Select ‘OK’ and the spectra will be tied as in the following:

In the above view the three DEPT experiments are tied together. This enables the user to align common peaks between the different 1D spectra and identify the multiplicities in the Carbon data. Also this view mode can be used for setting the common spectral reference between DEPT and Carbon data. One can also perform 1D peak picking and overlay the picked peaks on any related spectrum for comparison in this multiple spectral view.

Setting Reference and Appropriate Threshold

Open the DEPT ‘phasefile’ as a single spectrum, and leave at the first position for DEPT-45. Then, select ‘Peaks Table – 1D’ from the Display menu to display a blank 1D peaks table. Next, select ‘Load Table’ from the 1D Peaks Table, and select the c13.pks from the Q-2 directory. This will overlay the ¹³C peaks on the DEPT-45 spectrum. If the first peak from the left in the DEPT-45 spectrum matches with the peak symbol (+) of the ¹³C peak at 109.9134 ppm, then the reference is already set, and you can skip ahead to ‘Peak Picking of DEPT-45 Spectrum'.

However, if the DEPT-45 peak top is shifted from the ¹³C reference peak, then select ‘Set Reference’ from the Edit Menu, and place the cursor on the DEPT-45 peak. Then, keeping the left mouse button pressed (a circle around a ‘+’ symbol will appear) drag the cursor and release it at the center of the peak symbol (+) of the ¹³C peak at 109.9134 ppm. After releasing the left mouse button, the following dialog box will appear:

Click ‘OK’ and the reference of all the DEPT spectra will be the same as that of the ¹³C spectrum. Then, click 'Clear Table' from the 1D Peaks Table to remove the ¹³C peaks.

Peak Picking of DEPT-45 Spectrum

With the DEPT-45 spectrum still open, perform peak picking on the DEPT-45 spectrum by setting an appropriate threshold (i.e., 1.098e-2), and then selecting ‘Pick Peak Automatically - 1D’ from the Analysis menu. The program will pick 22 peaks. Next, sort the DEPT-45 peaks in the peaks table (descending order of X Value). Save the peaks as ‘dept45.pks’.

Peak Picking of DEPT-90 and DEPT-135 Spectra

Move the slider on the ‘1D Control Panel’ to the second position for DEPT-90. Then, set an appropriate threshold (i.e., 2.577e-2), and select ‘Pick Peak Automatically - 1D’ from the Analysis menu. The program will pick 6 peaks. Sort and save the peaks as ‘dept90.pks’.

Move the slider on the ‘1D Control Panel’ to the fourth position for DEPT-135. Then, set an appropriate threshold (i.e., 8.347e-3), and select ‘Pick Peak Automatically - 1D’ from the Analysis menu. Remember to check the ‘Negative Peaks’ option, since there are negative peaks in DEPT-135. The program will pick 22 peaks. Sort and save the peaks as ‘dept135.pks’.

I-5. Analysis of HMQC Spectrum.

The HMQC spectrum provides both C-H direct connectivity information and the ¹H chemical shifts of carbon-attached protons. Although the latter information can be obtained from 1D ¹H spectrum (if there aren’t many overlapping peaks), HMQC helps to resolve peak overlap and gives better separation for the ¹H peaks.

Setting Threshold

Open the HMQC spectrum by selecting ‘Open Spectrum’ from the File menu, and the spectrum will be displayed along with its ‘Threshold’ control palette as shown below:

The ‘Threshold’ palette contains the following features:

Auto Redraw (updates any changes made to the threshold palette immediately. Note: use only with small data sets that are sub-matrices of large 2D spectra, since it will slow down the refresh rate)

Threshold (defines threshold base by using the slider. Note that the first contour level corresponds to the threshold value.)

Separation (specifies the factor by which the base threshold is multiplied to determine the cutoff of each contour level)

Number of Levels (number of contour levels is by default 20 and can be modified in the *.ini files)

Starting Level (allows stepping through the available contour levels to resolve partially overlapped peaks and to locate peak tops)

On the ‘Threshold’ control palette, uncheck ‘Auto Redraw’, set the ‘Threshold’ as 3.548e-03, set the ‘Separation’ as 1.3, set the ‘Number of Levels’ as 20 and the ‘Starting Level’ as 1. Then, click ‘Update’ and the contours will be regenerated with the new values.

Setting Spectral Reference

SpecMan’s ‘Associate Reference Spectra’ option is useful for aligning 1D and 2D peaks, by allowing simultaneous display of 1D reference spectra on a 2D spectrum. Select ‘Associate Reference Spectra’ from the Display menu, and this will bring up the following:

Check the '1D Reference Spectra along X Axis' box and then select the accompanying 'Browse' button to select the ¹H ‘phasefile’ along the X-axis (F2 dimension).

Check the '1D Reference Spectra along Y Axis' and then select the accompanying 'Browse' button to select the ¹³C ‘phasefile’ along the Y-axis (F1 dimension). Check the '1D Reference Peak List along Y Axis' and then select the accompanying 'Browse' button to select the ‘13C.pks’ list along the Y-axis (F1 dimension). Additional details regarding this dialog box can be viewed by selecting the Help button from the dialog box.

After selecting the appropriate reference spectra and peaks table, click ‘OK’ to display the 1D reference spectrum and the grid lines drawn at the coordinates of the 1D ¹³C peaks (these will be used to verify the peak picking results later) as shown in the zoomed portion of the spectrum below:

To set reference on the HMQC spectrum, zoom on a well-resolved cross peak (for example, the cross peak at the lower left corner of the 2D spectrum). Check to see if both the 1D ¹H and ¹³C peaks are aligned with the center of this cross peak. If the 1D’s are not properly aligned, you must set reference on the HMQC spectrum.

To do so, select ‘Set Reference’ from the Edit Menu. Place the cross-hair cursor on the intersection of the HMQC cross peak. Then, keeping the left mouse button pressed (a circle around a ‘+’ symbol will appear), drag the cursor and release it at the intersection of the 1H and 13C peak coordinates. After releasing the left mouse button, the following dialog box will appear:

Click ‘OK’ and the new X and Y reference ppm values for that location will be accepted.

Correcting Chemical Shift Reference Offset between 1D and 2D

It is now important to verify the alignment of all 1D and 2D peak coordinates. Since the lower left peak has already been aligned, zoom on the HMBC cross peak at the upper right corner of the spectrum, and check to see if both the ¹H and ¹³C peaks are aligned well with the center of this HMBC cross peak. If there is a discrepancy, then the sweep width along X or Y needs to be adjusted.

To correct for sweep width, select ‘Spectral Parameters’ from the Edit Menu. Enter 5.9510 for the X Sweep Width and 219.1420 for the Y Sweep Width as shown below:

Then, click ‘OK’ to save the spectral parameters. SpecMan saves the modified parameters so that the parameters will be retained the next time the spectrum is accessed.

Automatic Peak Picking

To begin peak picking of the HMQC data, select ‘Pick Peaks Automatically – 2D’ from the Analysis menu. Make sure that the ‘Pick 2D Peaks’ dialog box is filled out as follows:

The ‘Merge Peak Multiplets’ option is utilized to pick the center of mass for cross peak multiplets, and can be performed in three separate modes:

Average

Weighted Average

Highest Peak

Using the ‘Merge Peak Multiplets’ with the ‘Weighted Average’ option, a multiplet will be merged and its center of mass will be used for the peak position. SpecMan’s peak picking algorithm uses peak width filters to discriminate noise from real peaks, and multiplets from independent peaks. These peak width filters are defined in terms of minimum and maximum box sizes for the search algorithm. For example, a peak with a width smaller than the minimum box size will be filtered out as a noise peak, while a cluster of peaks falling within the maximum box size will be merged as a multiplet.

If the user prefers to set these limits graphically, the user can select the ‘Set Graphically’ button for Peak Width Filter. This will momentarily close the ‘Pick 2D Peaks’ dialog box so that the user can draw a rectangular box (with the left mouse button) around a noise peak to set the minimum limit and around a cross peak for setting the maximum limit. During this process, the right mouse button and the four ‘zoom’ icons located at the top of the main window can be used to zoom in on the noise peaks and cross peaks used to set these limits.

However, for this HMQC example, the limits are known, and 0.005 and 0.05 are used as Minimum X and Minimum Y filters for filtering noise peaks, and 0.05 and 1.0 are used as Maximum X and Maximum Y filters for merging multiplets. After the dialog box has been filled in properly, click ‘OK’ and 33 peaks will be picked and displayed in the ‘2D Peaks Table’ as shown below:

Manual Editing of Peak Picking Results

For efficient structure elucidation, it is sometimes necessary to refine the automatic peak picking results. To analyze its peak picking results, SpecMan allows the user to examine each 2D peak (one at a time) in a zoomed mode. To do this, zoom in on any peak of interest (so that it is the only peak visible in the spectral window), and then click on Peak #1 in the ‘2D Peaks Table.’ This will then display Peak #1 at the same level of zoom as the initially zoomed peak. Now examine the zoomed peak and compare its center with the intersection of the grid lines to make sure that the peak has been correctly picked. Then, use your keyboard’s up and down arrow keys to step through and examine the remaining peaks in the ‘2D Peaks Table‘ at the same zoom level.

In this HMQC spectrum, a few peaks were not picked at the real center of their multiplets because of their line shapes (7, 11, 12, 17, 23, 24, and 26). To correct this, select ‘Move Peaks’ from the Analysis menu, left mouse click on the peak symbol (+) and drag it to a new location. This new information will be updated in the 2D Peaks Table. Finally, 33 peaks are retained.

Sorting and Saving Peaks, Extracting ¹H Chemical Shifts

Click the ‘Edit Table’ button from the ‘2D Peaks Table.’ The following ‘Edit Peaks Table’ dialog box will appear:

Check the following options:

‘Sort Table Entries’

‘Descending’

‘Y Value’ (¹³C chemical shifts)

‘Renumber ID's’

‘Remove Redundant Peaks’

‘Extract Coordinates to 1D Table -X’

This last option (Extract Coordinates to 1D Table – X) will extract the ¹H peak information along the X dimension from the HMQC spectrum. Click ‘OK’ and 33 ¹H peaks will be displayed in the 1D Peaks Table. Sort the ¹H peaks in the 1D Peaks Table by selecting ‘Sort Table Entries,’ ‘Descending - X Value’, and ‘Renumber Table ID's.’ Then, select ‘Save’ from the 1D Peaks Table and save the peaks as ‘h1.pks.’ Also, save the 2D HMQC peaks as ‘hmqc.pks.’

Now that the ¹H chemical shifts have been extracted from the HMQC spectrum, they can be displayed as grid lines on the HMQC spectrum. To do this, select ‘Associate Reference Spectra’ from the Display menu, click the browse button following the 1D Reference Peak List along X, and choose ‘h1.pks.’ Click ‘OK’ and the ¹H grid lines will be displayed along with the previously associated ¹³C grid lines, as shown below:

Together with the ¹³C grid lines, the user can verify the peak picking results again. It is important to extract the ¹H chemical shifts as accurately as possible because they will be used for the peak picking of the other 2D spectra.

Peaks Due to Protons Attached to Heteroatoms

The ¹H peak list (h1.pks) extracted from the HMQC spectral data does not include any information from the protons attached to heteroatoms. Please note that the OH peaks are hardly visible, and there won’t be any addition of proton peaks due to heteroatoms. However, if the user wants to manually add this information, the user should open and display the ¹H spectrum, and load the ¹H peaks file by selecting ‘Load Table’ from the 1D Peaks Table (choose ‘Peaks Tables - 1D’ from the Analysis menu if the 1D Peaks Table is not already open). The ¹H peaks will then be displayed on the ¹H spectrum.

The unlabelled peaks that correspond to the protons attached to heteroatoms can be added by selecting ‘Add Peaks Manually - Without Refine’ from the Analysis menu. Then, move the cursor to the location of the unlabelled peak, and click the left mouse button to add a new peak at that location. Repeat this step until all unlabelled peaks have been added. Reselect ‘Add Peaks Manually - Without Refine’ from the Analysis menu to deactivate this option.

I-6. Analysis of DQF-COSY Spectrum

Setting Spectral Reference and Threshold

Open the DQF-COSY (‘COSY’) spectral file ‘phasefile’ from the COSY folder and set the following ‘Threshold’ palette controls:

‘Threshold’ = 7.546e-03

‘Separation’ = 1.4

‘Number of Levels’ = 24

The spectrum will appear as follows:

It is recommended to use a fairly low threshold so that all of the weak COSY peaks will appear. It is important that all of the COSY peaks, including the very weak peaks are picked, since NMR-SAMS uses the negative information from COSY. For example, two proton-bearing carbons can be forbidden to connect if their protons show no COSY peaks.

Grid Intelligence-based Peak Picking

Select ‘Associate Reference Spectra’ from the Display menu to display the ¹H reference spectrum and the ¹H peaks list (h1.pks) along both the X-axis and the Y-axis. Next, the spectral reference is calibrated in the same manner as the HMQC spectrum by checking the alignment of the 1D reference peaks with the 2D peaks.

The COSY spectrum peaks are picked using a novel peak picking method called grid intelligence-based peak picking. In this procedure, the grid intersection points (instead of the minimum peak width) are used as filters in the search algorithm for peak picking. After a multiplet is merged and its center of mass is calculated, SpecMan attempts to locate a grid intersection point within the specified minimum tolerance (determined by the user). If such a grid intersection point is found, the peak center is retained as a real peak. Otherwise, it is rejected.

To pick the COSY peaks using grid intelligence-based peak picking, select ‘Pick Peaks Automatically - 2D’ from the Analysis menu. This will bring up the ‘Pick 2D Peaks’ dialog box. Check the following options:

‘Positive Peaks’

‘Negative Peaks’

‘Grid Intelligence’

Grid Distance Filter’s Minimum X ppm = 0.008

Grid Distance Filter’s Minimum Y ppm = 0.008

‘Merge Peak Multiplets’ = Weighted Average

Peak Width Filter’s Minimum X and Y ppm = 0.005 (filters for noise peaks)

Peak Width Filter’s Maximum X and Y ppm = 0.05 (filters for considering a cross-peak as a multiplet)

Next, click ‘OK’ and 86 peaks will be picked in less than 1 minute.

It is not necessary to discard the COSY diagonal peaks, since NMR-SAMS will automatically disregard them later. In addition, NMR-SAMS will also merge symmetry-related peaks for all homonuclear spectra, including COSY. Although it is not crucial to pick both sets of symmetric peaks around the diagonal, SpecMan does pick both sets of symmetric peaks since it can be beneficial if peaks on one side of the diagonal are cleaner than the symmetric peaks on the other side of the diagonal. In cases where multiple grid line intersections are close to a peak and it is difficult to resolve as an unambiguous correlation, it is best advised to leave the picked peak as it is. NMR-SAMS will automatically include the different possible correlations to 1D peaks, and treat the cross peak as ambiguous correlation information.

In this COSY example, some noise peaks are removed, some ignored peaks are added (the peaks around (1.42, 1.20) and (1.37, 1.20)), and some incorrectly picked peaks that are located away from grid line intersections are corrected. Finally, 87 peaks are retained (Please note that if you have any trouble cleaning the COSY peaks, please load the ‘cosy.pks’ file from the sample data into the 2D Peaks Table to see which peaks were removed, added, or modified). The 2D Peaks Table is sorted and renumbered using the ‘Edit Table’ option, and then saved as ‘cosy.pks’ using the ‘Save Table’ option.

Please note that the peak picking results depend upon chemical shift reference and alignment between the 1D and 2D spectra (when grid intelligence is used), in addition to the parameters in the ‘Pick 2D Peaks’ dialog box. If any of these dependencies are modified, the user may get slightly different results, but this should not drastically affect NMR-SAMS’ subsequent structure elucidation.

I-7. Analysis of HMBC Spectrum

Setting Spectral Reference and Threshold

Open the HMBC spectral file ‘phasefile’ from the HMBC folder and set the following ‘Threshold’ palette controls:

‘Threshold’ = 3.578e-03

‘Separation’ = 1.2

‘Number of Levels’ = 20

The spectrum will appear as follows:

Select ‘Associate Reference Spectra’ from the Display menu to display the ¹H reference spectrum and the ¹H peaks list (h1.pks) along the X-axis and the ¹³C reference spectrum and the ¹³C peaks list (c13.pks) along the Y-axis. Next, the spectral reference is calibrated in the same manner as the HMQC spectrum by checking the alignment of the 1D reference peaks with the 2D peaks.

Grid Intelligence-based Peak Picking

Similar to the DQF-COSY spectrum, the peaks of HMBC are picked utilizing SpecMan’s grid intelligence-based option. To pick the HMBC peaks using grid intelligence-based peak picking, select ‘Pick Peaks Automatically - 2D’ from the Analysis menu. This will bring up the ‘Pick 2D Peaks’ dialog box. Check the following options:

‘Positive Peaks’

‘Grid Intelligence’

Grid Distance Filter’s Minimum X ppm = 0.011

Grid Distance Filter’s Minimum Y ppm = 0.11

‘Merge Peak Multiplets’ = Weighted Average

Peak Width Filter’s Minimum X ppm = 0.003 Minimum Y ppm = 0.05

Peak Width Filter’s Maximum X ppm = 0.03 Maximum Y ppm = 0.5

Next, click ‘OK’ and 150 peaks will be picked in less than 1 minute.

With regards to analysis of the peak picking results, the display of the reference 1D spectra as well as the grid lines on the 2D spectrum facilitates the user’s verification of the peak picking results to a great extent. In addition, certain areas of the spectrum can be looked at in more detail. For example, in the t₁-ridge area, the ‘Starting Level’ value in the ‘Threshold’ palette can be increased to get a better display of the real peaks. In cases where multiple grid line intersections are close to a peak and it is difficult to resolve as an unambiguous correlation, it is best advised to leave the picked peak as it is. NMR-SAMS will automatically include the different possible correlations to 1D peaks, and treat the cross peak as ambiguous correlation information. Finally, 133 peaks are retained (Please note that it you have any trouble cleaning the HMBC peaks, please load the ‘hmbc.pks’ file from the sample data into the 2D Peaks Table to see which peaks were removed, added, or modified). The 2D Peaks Table is sorted and renumbered using the ‘Edit Table’ option, and then saved as ‘hmbc.pks’ using the ‘Save Table’ option.

I-8. Analysis of NOESY Spectrum

Setting Spectral Reference and Threshold

Open the NOESY spectral file ‘phasefile’ from the NOESY folder and set the following ‘Threshold’ palette controls:

‘Threshold’ = 4.235e-03

‘Separation’ = 1.2

‘Number of Levels’ = 20

(Make sure to uncheck the ‘Negative Levels’)

The spectrum will appear as follows:

Grid Intelligence based Peak Picking

Similar to the HMBC spectrum, the peaks of NOESY are picked utilizing SpecMan’s grid intelligence-based option. To pick the NOESY peaks using grid intelligence-based peak picking, select ‘Pick Peaks Automatically - 2D’ from the Analysis menu. This will bring up the ‘Pick 2D Peaks’ dialog box. Check the following options:

‘Positive Peaks’

‘Grid Intelligence’

Grid Distance Filter’s Minimum X ppm = 0.008

Grid Distance Filter’s Minimum Y ppm = 0.008

‘Merge Peak Multiplets’ = Weighted Average

Peak Width Filter’s Minimum X ppm = 0.01 Minimum Y ppm = 0.01

Peak Width Filter’s Maximum X ppm = 0.05 Maximum Y ppm = 0.05

Next, click ‘OK’ and 68 peaks will be picked in less than 1 minute.

Please note that the NOESY data is utilized by NMR-SAMS only when the user opts to use the negative information of COSY together with NOESY. For example, if there is neither a COSY nor a NOESY peak observed between two proton-attached carbon atoms, then this carbon pair is forbidden to connect. In this Q-2 example, the NOESY data will not be utilized in the structure elucidation process, but will be used for creating and exporting the NOE assignments after the structure has been generated (for details see Section II-15). For this reason, the NOESY peaks table is not manually revised now.

Part II Computer-Assisted Structure Elucidation with NMR-SAMS

II-1. Introduction

The first step in NMR-SAMS is to convert the ¹H, ¹³C, DEPT-45, DEPT-90, DEPT-135, DQF-COSY, HMQC, HMBC, and NOESY peak tables from SpecMan into correlation information for NMR-SAMS. Such correlation information is then interpreted to define bond constraints on the atoms labeled by the 1D chemical shifts. These bond constraints are transformed into a set of mutually consistent C-C bond constraints, and based on this information, a set of structural building blocks are generated and an atom-atom connection matrix (ACMX) is set up. Finally, possible 2D structures are generated. If complete 2D structures are not obtained during structure elucidation, then NMR-SAMS reports only the largest possible partial structure with resonance assignments.

During the 2D structure generation process, NMR-SAMS also provides complete resonance assignments consistent with NMR data for the candidate 2D structures. Alternatively, NMR-SAMS can provide resonance assignments for 2D structures that are proposed as possible target structures by the user.

II-2. Getting Started with NMR-SAMS

From the Program Manager or the Start Menu, click the NMR-SAMS icon in the Spectrum2001 group to launch the NMR-SAMS program. Once the program has been initiated, the main NMR-SAMS window appears, as well as a ‘Status Window’ that lists the current status of the structure elucidation process and also suggests the next possible steps for the user to take.

For this part of the tutorial, please use the SpecMan-generated peak lists located in the following directory: C:\Spectrum2001\Data\NMR-SAMS\Q-2.

II-3. Opening New Working Data Set

Select ‘New’ from the File menu, and an ‘Open File’ dialog box will appear. Click on the ‘q-2.mdf’ file and then modify its root name in the ‘File name’ text box to ‘q-2test.mdf’ as shown below:

Then, click ‘OK’ and NMR-SAMS will create the following files in the current directory:

q-2test.mdf – master data file. Stores all of the intermediate and final results.

q-2test.par – default parameter file. Stores all of the control parameters used for data interpretation and structure generation. NMR-SAMS uses a set of default control parameters, but also provides the user the option of changing these control parameters (To do so, select ‘Parameters - NMR Interpretation, Setting up ACMX, or 2D Structure Generation’ from the Edit menu).

q-2test.nmr – NMR data file. Stores all of the SpecMan-converted peaks table data.

The user can edit this file by selecting, ‘NMR Data File’ from the Edit menu.

q-2test.log – log file. Stores all of the warning and error messages produced during the analysis. The user can view the log file by selecting ‘Log File’ from the Edit menu.

q-2test.str – Stores the connection table of the generated structures and their resonance assignments.

q-2test.lock – Prevents the data set from being opened simultaneously by two users.

Next, enter the molecular formula of Q-2 (C₃₀H₄₈O₃) into the ‘Input Molecular Formula’ dialog box as shown below:

Then click, ‘OK’ and the molecular formula is automatically interpreted for element composition and common valences. This information is written into the q-2test.mdf file after the keyword, ‘Atoms…’. If an atom has an unusual valence, specify the valence after the element symbol (C47H51N(V)O14 where N(V) indicates a nitrogen atom with a valence of 5). Otherwise, the common valence will be adopted.

Please note that if the molecular formula is unknown, the user can enter ‘unknown’ (please see section II-18 for instructions regarding structure elucidation with an unknown molecular formula).

II-4. Conversion of SpecMan ¹H Peak List.

Next, select ‘Create NMR Data File - H-1’ from the File menu, and NMR-SAMS will display a ‘Convert SpecMan H-1 Peaks Table’ dialog box. Select the ‘h1.pks’ file created in SpecMan as shown below:

Click ‘OK’ and the following dialog box will appear showing that 33 ¹H peaks were converted and written into the q-2test.nmr file:

Click ‘OK’ and the following information dialog box will appear asking the user to supply the multiplicity information for the ¹H peaks.

Click ‘OK’ to accept this message. Note that NMR-SAMS recognizes only the following multiplet options: singlet (s), doublet (d), triplet (t), and quartet (q). All other multiplicities are entered either as unknown (u) or multiplet (m). Multiplet information will be used to eliminate inappropriate bonds during structure generation.

Select ‘NMR Data File’ from the Edit menu to modify the information obtained from the ¹H peaks in the ‘q-2test.nmr’ file. Edit this file by replacing the unknown multiplicity ‘u’ of 8 peaks (1, 2, 12, 25, 28, 29, 30 and 32) with singlet multiplicity ‘s’, as shown in Table I:

Table I. The ¹H peak list of Q-2

H1: C:\Spectrum2001\Data\NMR-SAMS\Q-2\h1.pks

#1. 4.930 s ;1

#2. 4.755 s ;2

#3. 3.509 u ;3

#4. 3.435 u ;4

#5. 2.725 u ;5

#6. 2.611 u ;6

#7. 2.235 u ;7

#8. 2.232 u ;8

#9. 1.924 u ;9

#10. 1.863 u ;10

#11. 1.830 u ;11

#12. 1.778 s ;12

#13. 1.752 u ;13

#14. 1.653 u ;14

#15. 1.565 u ;15

#16. 1.544 u ;16

#17. 1.540 u ;17

#18. 1.513 u ;18

#19. 1.439 u ;19

#20. 1.419 u ;20

#21. 1.370 u ;21

#22. 1.369 u ;22

#23. 1.368 u ;23

#24. 1.249 u ;24

#25. 1.208 s ;25

#26. 1.196 u ;26

#27. 1.196 u ;27

#28. 1.062 s ;28

#29. 1.051 s ;29

#30. 0.993 s ;30

#31. 0.969 u ;31

#32. 0.818 s ;32

#33. 0.811 u ;33

The line beginning with ‘H1’ indicates the start of ¹H peak list. The entries in the following lines indicate Peak ID, chemical shift, multiplicity, and comments (optional) for each ¹H peak. In this table, the number in the comment field corresponds to the ID of the peak in the SpecMan peaks table. Comments are not used by NMR-SAMS.

II-5. Conversion of SpecMan ¹³C and DEPT Peak List.

Next, select ‘Create NMR Data File – C13 and DEPT’ from the File menu, and NMR-SAMS will display a ‘Convert SpecMan C13 and DEPT Peaks Table’ dialog box. Select ‘Browse’ to locate the ‘c13.pks,’ ‘dept90.pks,’ and dept135.pks’ files created in SpecMan as shown below:

The default value of 0.08 is used for ‘Tolerance Along X’. Click ‘OK’ and the following dialog box will appear showing that 29 ¹³C peaks were converted and written into the q-2test.nmr file:

Click ‘OK’ and the following information dialog box will appear saying that the program has compared the number of peaks with the molecular formula and found that there are fewer ¹³C peaks than expected (since the molecular formula has 30 Carbons):

Click ‘OK’ to accept this warning message, and then select ‘NMR Data File’ from the Edit menu to modify the information obtained from the ¹³C, DEPT-90 and DEPT-135 peaks in the ‘q-2test.nmr’ file. From the ¹³C peak height, as well as the number of corresponding cross peaks in the HMQC spectrum, it is easy to identify ¹³C peak #27 (16.441ppm, q) as a degenerate one. Edit this file by adding a new peak to the end of the ¹³C peak list. To do this, copy peak #27, and paste it to the end of the peak list. Then, renumber it to peak #30 and modify the chemical shift from 16.441 to 16.442. We can use the comment section of the line (after the semicolon) to identify this as a split peak for our information later (the remaining peaks are left unchanged). These changes are displayed in Table II:

Table II. The ¹³C peak list of Q-2

C13: C:\Spectrum2001\Data\NMR-SAMS\Q-2\c13.pks

#1. 178.822 s ;1

#2. 151.323 s ;2

#3. 109.931 t ;3

#4. 78.147 d ;4

#5. 56.647 s ;5

#6. 55.956 d ;6

#7. 50.997 d ;7

#8. 49.814 d ;8

#9. 47.783 d ;9

#10. 42.877 s ;10

#11. 41.151 s ;11

#12. 39.540 s ;12

#13. 39.318 t ;13

#14. 38.648 d ;14

#15. 37.596 t ;15

#16. 37.556 s ;16

#17. 34.868 t ;17

#18. 32.901 t ;18

#19. 31.246 t ;19

#20. 30.301 t ;20

#21. 28.679 q ;21

#22. 28.330 t ;22

#23. 26.149 t ;23

#24. 21.243 t ;24

#25. 19.507 q ;25

#26. 18.813 t ;26

#27. 16.441 q ;27

#28. 16.340 q ;28

#29. 14.929 q ;29

#30. 16.442 q ;27 Split from #27

The line beginning with ‘C13’ indicates the start of ¹³C peak list. The entries in the following lines indicate Peak ID, chemical shift, multiplicity, and comments (optional) for each ¹³C peak. In this table, the number in the comment field corresponds to the ID of the peak in the SpecMan peaks table. Comments are not used by NMR-SAMS.

II-6. Conversion of SpecMan COSY Peak List.

Next, select ‘Create NMR Data File – COSY’ from the File menu, and NMR-SAMS will display a ‘Convert SpecMan COSY Peaks Table’ dialog box. Select ‘Browse’ to locate the ‘cosy.pks’ file created in SpecMan as shown below:

The default value of 0.005 is used for ‘Tolerance Along X’ and ‘Tolerance Along Y’. Click ‘OK’ and the following warning box will appear cautioning the user about ambiguous correlations for some peaks:

Click ‘OK to All’ to accept this warning message, and the following dialog box will appear showing that 32 COSY peaks were converted and written into the q-2test.nmr file:

The user is also notified that since NMR-SAMS assumes that all peaks are short-range couplings with unknown J-coupling constants, the user needs to denote some of the potential long-range coupled peaks as "weak" ones. The user can also supply accurate J-coupling constants, and let the program determine nature of coupling (long-range or short-range) based on that information.

Click ‘OK’ to accept this message, and then select ‘NMR Data File’ from the Edit menu to modify the information obtained from the COSY peaks in the ‘q-2test.nmr’ file. From their weak intensities due to potential long-range coupling, six COSY peaks are identified for editing. Modify the intensity level of these 6 peaks (1, 2, 3, 8, 20 and 29) from the default value of ‘3’ (strong) to ‘1’ (weak), as shown in Table III.

(Please note that a potential long-range coupling COSY peak is usually interpreted as 3-5 intervening bonds between the correlated protons, which also covers the possibility of vicinal coupling. A short-range coupling is usually interpreted as 2-3 intervening bonds between the correlated protons. If a long-range coupling is mistakenly interpreted as a short-range coupling, NMR-SAMS will not generate the correct structure. So it is very important to denote these potential long-range coupling peaks. Geminal coupling is always automatically detected by the program.)

Table III. The COSY peak list of Q-2

COSY: C:\Spectrum2001\Data\NMR-SAMS\Q-2\cosy.pks

#1. (1 - 2) 1 0.00 ;1+4

#2. (1 - 12) 1 0.00 ;2+31

#3. (2 - 12) 1 0.00 ;3+32

#4. (3 - 7 8) 3 0.00 ;6+18

#5. (3 - 13) 3 0.00 ;7+33

#6. (3 - 18) 3 0.00 ;5+49

#7. (4 - 11) 3 0.00 ;8+30

#8. (5 - 9) 1 0.00 ;12+21

#9. (5 - 13) 3 0.00 ;11+34

#10. (5 - 26 27) 3 0.00 ;82+87

#11. (6 - 10) 3 0.00 ;15+26

#12. (6 - 16 17) 3 0.00 ;14+83

#13. (6 - 24) 3 0.00 ;13

#14. (7 8 - 15) 3 0.00 ;17+38

#15. (7 8 - 18) 3 0.00 ;16+48

#16. (9 - 20) 3 0.00 ;22

#17. (9 - 26 27) 3 0.00 ;20

#18. (10 - 16 17) 3 0.00 ;27+47

#19. (10 - 24) 3 0.00 ;24

#20. (10 - 28) 1 0.00 ;23

#21. (11 - 14) 3 0.00 ;29+36

#22. (11 - 31) 3 0.00 ;28

#23. (14 - 31) 3 0.00 ;37+75

#24. (16 17 - 19) 3 0.00 ;51+85

#25. (16 17 - 21 22 23) 3 0.00 ;43+55

#26. (16 17 - 24) 3 0.00 ;45+60

#27. (16 17 - 33) 3 0.00 ;84

#28. (19 - 21 22 23) 3 0.00 ;52+58

#29. (19 - 29) 1 0.00 ;50

#30. (20 - 26 27) 3 0.00 ;53+64

#31. (21 22 23 - 26 27) 3 0.00 ;56+71

#32. (21 22 23 - 33) 3 0.00 ;86

The entries in the lines above for each COSY cross peak indicate Cross Peak ID, ID’s of the correlated ¹H peaks (in parenthesis; for ambiguous correlations the ID’s of all possible ¹H peak correlations are included), peak intensity levels (which are classified as four types; strong (3), medium (2), weak (1), and unknown (0) – the default value is 3), J-coupling (optional, the default value is 0), and comments (optional, with a maximum size of 80 characters).

For short-range coupled DQF-COSY peaks, intensity levels should be either 3 or 2. For long-range coupled DQF-COSY peaks, the intensity level should be 1. If an intensity level of 0 is used, the program will expect actual J-coupling values in the field that represents J-coupling. One or more spaces are used as a delimiter for all items, except for comments which are separated by “;”. Items marked as optional can be omitted unless an item following it is included. In such a case, please include default values for ignored items even if they don’t get used. Comments can always be included as long as they follow a “;”. In this table, the numbers in the comment field correspond to the ID’s of the peaks in the SpecMan peaks table. For merged peaks, these numbers are shown with a + sign. Comments are not currently used by NMR-SAMS.

II-7. Conversion of SpecMan HMQC Peak List.

Next, select ‘Create NMR Data File – HMQC’ from the File menu, and NMR-SAMS will display a ‘Convert SpecMan HMQC Peaks Table’ dialog box. Select ‘Browse’ to locate the ‘hmqc.pks’ file created in SpecMan as shown below:

The default value of 0.005 is used for ‘Tolerance Along X’ and the ‘Tolerance Along Y’ value is changed to 0.1. Click ‘OK’ and the following warning box will appear cautioning the user about a ¹³C peak with fewer correlations than expected:

Click ‘OK to All’ to accept this warning message, and 33 HMQC peaks are converted and written into the q-2test.nmr file as shown in Table IV:

(Please note that unlike other 2D peaks where ambiguous correlations are allowed, HMQC peaks must have exactly two correlated 1D peaks).

Table IV. The HMQC peak list of Q-2

HMQC: C:\Spectrum2001\Data\NMR-SAMS\Q-2\hmqc.pks

#1. (3 - 1) ;2

#2. (3 - 2) ;1

#3. (4 - 4) ;3

#4. (6 - 33) ;4

#5. (7 - 21) ;5

#6. (8 - 13) ;6

#7. (9 - 3) ;7

#8. (13 - 14) ;8

#9. (13 - 31) ;9

#10. (14 - 5) ;10

#11. (15 - 7) ;12

#12. (15 - 15) ;11

#13. (17 - 19) ;14

#14. (17 - 23) ;13

#15. (18 - 6) ;16

#16. (18 - 17) ;15

#17. (19 - 8) ;18

#18. (19 - 18) ;17

#19. (20 - 10) ;19

#20. (20 - 24) ;20

#21. (21 - 25) ;21

#22. (22 - 11) ;22

#23. (23 - 9) ;24

#24. (23 - 26) ;23

#25. (24 - 20) ;25

#26. (24 - 27) ;26

#27. (25 - 12) ;27

#28. (26 - 16) ;28

#29. (26 - 22) ;29

#30. (27 - 32) ;31

#31. (28 - 30) ;32

#32. (29 - 28) ;33

#33. (30 - 29) ;30

The line beginning with ‘HMQC’ indicates the start of HMQC peak list. After the keyword ‘HMQC’, following a blank space, comments may be added up to 80 characters in length. The entries in the rest of the lines represent the Cross Peak ID, ID’s of the correlated ¹³C & ¹H peaks shown in parenthesis, and comments (optional, with a maximum size of 80 characters), for each HMQC cross peak. One or more spaces are used as a delimiter for all items except comments which are separated by “;”. Comments are not currently used by NMR-SAMS.

II-8. Conversion of SpecMan HMBC Peak List.

Next, select ‘Create NMR Data File – HMBC’ from the File menu, and NMR-SAMS will display a ‘Convert SpecMan HMBC Peaks Table’ dialog box. Select ‘Browse’ to locate the ‘hmbc.pks’ file created in SpecMan as shown below:

The default value of 0.005 is used for ‘Tolerance Along X’ and the default value of 0.08 is used for ‘Tolerance Along Y’. Click ‘OK’ and the following warning box will appear indicating that ambiguous cross peaks have been obtained:

Click ‘OK to All’ to ignore this message (and to ignore similar messages for this example).

The following warning message will appear to mention that a SpecMan peak has been discarded because its X (¹H) or Y (¹³C) chemical shift does not match any 1D peak (This happens if it is an artifact or if the center of a peak is not accurately located):

Click ‘OK to All’ to ignore this message, and 129 HMBC peaks are converted and written into the q-2test.nmr file as shown in Table V:

(Please note that it can be seen that some peaks have ambiguous correlation to 1D peaks due to very close chemical shifts. Although NMR-SAMS can use ambiguous correlation information, too many ambiguous correlations will undermine the efficiency of the subsequent structure generation. So whenever possible the user should resolve ambiguities manually. In this example, there are 36 ambiguous cross peaks that are not well resolved).

Table V. The HMBC peak list of Q-2

HMBC: C:\Spectrum2001\Data\NMR-SAMS\Q-2\hmbc.pks

#1. (1 - 6) ;3

#2. (1 - 7 8) ;4

#3. (1 - 13) ;5

#4. (1 - 15) ;2

#5. (1 - 16 17) ;1

#6. (2 - 1) ;9

#7. (2 - 2) ;11

#8. (2 - 3) ;8

#9. (2 - 8) ;6

#10. (2 - 12) ;7

#11. (2 - 18) ;10

#12. (3 - 3) ;12

#13. (3 - 12) ;13

#14. (4 - 11) ;14

#15. (4 - 14) ;16

#16. (4 - 25) ;17

#17. (4 - 30) ;15

#18. (5 - 5) ;25

#19. (5 - 6) ;20

#20. (5 - 7 8) ;26

#21. (5 - 10) ;24

#22. (5 - 13) ;21

#23. (5 - 15) ;22

#24. (5 - 16 17) ;23

#25. (5 - 18) ;19

#26. (5 - 24) ;18

#27. (6 - 14) ;30

#28. (6 - 19) ;32

#29. (6 - 21 22 23) ;27

#30. (6 - 25) ;31

#31. (6 - 30) ;29

#32. (6 - 32) ;28

#33. (7 - 9) ;33

#34. (7 - 20) ;38

#35. (7 - 21 22 23) ;37

#36. (7 - 29) ;35

#37. (7 - 31) ;36

#38. (7 - 32) ;34

#39. (8 - 3) ;43

#40. (8 - 5) ;41

#41. (8 - 6) ;42

#42. (8 - 7 8) ;40

#43. (8 - 16 17) ;39

#44. (9 - 1) ;45

#45. (9 - 2) ;44

#46. (9 - 7 8) ;46

#47. (9 - 12) ;49

#48. (9 - 13) ;48

#49. (9 - 15) ;47

#50. (9 - 18) ;50

#51. 10 - 5) ;57

#52. (10 - 6) ;54

#53. 10 - 9) ;55

#54. (10 - 10) ;56

#55. (10 - 13) ;59

#56. (10 - 19) ;51

#57. (10 - 24) ;52

#58. (10 - 28) ;53

#59. (10 - 29) ;58

#60. (11 - 10) ;65

#61. (11 - 16 17) ;63

#62. (11 - 19) ;60

#63. (11 - 20) ;64

#64. (11 - 21 22 23) ;61

#65. (11 - 28) ;66

#66. (11 - 29) ;62

#67. (12 - 4) ;69

#68. (12 - 11) ;67

#69. (12 - 16 17) ;72

#70. (12 - 25) ;70

#71. (12 - 30) ;68

#72. (12 - 33) ;71

#73. (14 - 9) ;79

#74. (14 - 13) ;78

#75. (14 - 20) ;76

#76. (14 - 24) ;75

#77. (14 - 26 27) ;80

#78. (14 - 28) ;77

#79. (15 16 - 7 8) ;81

#80. (15 16 - 11) ;89

#81. (15 16 - 13) ;90

#82. (15 16 - 14) ;87

#83. (15 16 - 16 17) ;88

#84. (15 16 - 18) ;84

#85. (15 16 - 21 22 23) ;85

#86. (15 16 - 31) ;83

#87. (15 16 - 32) ;82

#88. (15 16 - 33) ;86

#89. (17 - 16 17) ;92

#90. (17 - 21 22 23) ;91

#91. (17 - 29) ;93

#92. (17 - 33) ;94

#93. (18 - 10) ;97

#94. (18 - 13) ;98

#95. (18 - 15) ;96

#96. (18 - 24) ;95

#97. (19 - 3) ;99

#98. (19 - 7 8) ;101

#99. (19 - 15) ;100

#100. (20 - 6) ;104

#101. (20 - 16 17) ;102

#102. (20 - 28) ;103

#103. (21 - 4) ;106

#104. (21 - 30) ;105

#105. (21 - 33) ;107

#106. (22 - 14) ;109

#107. (22 - 31) ;108

#108. (23 - 5) ;111

#109. (23 - 13) ;112

#110. (23 - 26 27) ;110

#111. (24 - 5) ;116

#112. (24 - 21 22 23) ;115

#113. (24 - 26 27) ;114

#114. (25 - 1) ;120

#115. (25 - 2) ;119

#116. (25 - 3) ;118

#117. (26 - 19) ;122

#118. (26 - 21 22 23) ;123

#119. (26 - 32 33) ;121

#120. (27 28 30 - 4) ;130

#121. (27 28 30 - 25) ;129

#122. (27 28 30 - 32 33) ;128

#123. (27 30 - 14) ;126

#124. (27 30 - 19) ;124

#125. (27 30 - 21 22 23) ;127

#126. (27 30 - 31) ;125

#127. (29 - 5) ;132

#128. (29 - 10) ;133

#129. (29 - 24) ;131

The line beginning with ‘HMBC’ indicates the start of HMBC peak list. After the keyword ‘HMBC’, following a blank space, comments may be added up to 80 characters in length. The entries in the rest of the lines represent the Cross Peak ID, ID’s of the correlated ¹³C & ¹H peaks shown in parenthesis, and comments (optional, with a maximum size of 80 characters), for each HMBC cross peak. One or more spaces are used as a delimiter for all items except comments which are separated by “;”. Comments are not currently used by NMR-SAMS.

II-9. Conversion of SpecMan NOESY Peak List

Next, select ‘Create NMR Data File – NOESY’ from the File menu, and NMR-SAMS will display a ‘Convert SpecMan NOESY Peaks Table’ dialog box. Select ‘Browse’ to locate the ‘noesy.pks’ file created in SpecMan.

The default value of 0.005 is used for ‘Tolerance Along X’ and ‘Tolerance Along Y’. Click ‘OK to All’ to the resulting warning messages, and finally, 49 NOESY peaks are converted and written into the q-2test.nmr file as shown in Table VI

Table VI. The NOESY peak list of Q-2

NOESY: C:\Spectrum2001\Data\NMR-SAMS\Q-2\noesy.pks

#1. (1 - 2) 3 0.16 ;1+2

#2. (1 - 3) 3 0.01 ;4

#3. (1 - 12) 3 0.01 ;19

#4. (2 - 12) 3 0.01 ;20

#5. (3 - 7 8) 3 0.01 ;13

#6. (4 - 25) 3 0.01 ;60

#7. (4 - 32) 3 0.01 ;88

#8. (5 - 29) 3 0.02 ;75

#9. (6 - 16) 3 0.02 ;9

#10. (6 - 17) 3 0.01 ;33

#11. (7 8 - 15) 3 0.05 ;11+27

#12. (7 8 - 18) 3 0.04 ;35

#13. (9 - 20) 3 0.01 ;43

#14. (9 - 26 27) 3 0.03 ;14

#15. (10 - 24) 3 0.07 ;15+56

#16. (10 - 29) 3 0.02 ;74

#17. (11 - 14) 3 0.01 ;25

#18. (11 - 32) 3 0.02 ;87

#19. (12 - 25) 3 0.01 ;62

#20. (13 - 16 17) 3 0.01 ;32

#21. (13 - 28) 3 0.01 ;21

#22. (14 - 20) 3 0.01 ;24+42

#23. (14 - 31) 3 0.01 ;23

#24. (14 - 32) 3 0.01 ;89

#25. (15 - 21 22 23) 3 0.01 ;26

#26. (16 - 21 22 23) 3 -0.02 ;51

#27. (16 17 - 25) 3 0.01 ;59

#28. (16 17 - 28) 3 0.01 ;71

#29. (16 17 - 32 33) 3 0.01 ;91

#30. (19 - 21 22 23) 3 0.04 ;52

#31. (19 - 28) 3 0.02 ;69

#32. (19 - 32) 3 0.01 ;84

#33. (20 - 26 27) 3 0.07 ;40+66

#34. (20 - 32) 3 0.01 ;85

#35. (21 - 28) 3 0.02 ;45

#36. (21 22 - 30) 3 0.01 ;78

#37. (21 22 23 - 24) 3 0.01 ;50

#38. (21 22 23 - 30) 3 0.02 ;48

#39. (21 22 23 - 31) 3 0.01 ;47

#40. (21 22 23 - 32 33) 3 0.02 ;46

#41. (22 23 - 33) 3 0.02 ;93

#42. (24 - 28) 3 0.01 ;70

#43. (25 - 30) 3 0.03 ;77

#44. 25 - 32) 3 0.03 ;86

#45. (25 - 32 33) 3 0.01 ;63

#46. (26 27 - 28) 3 0.02 ;68

#47. (29 - 32) 3 0.04 ;83

#48. (30 - 32) 3 0.09 ;79+90

#49. (31 - 33) 3 0.01 ;94

The line beginning with ‘NOESY’ indicates the start of the NOESY peak list. After the keyword ‘NOESY’, following a blank space, comments may be added up to 80 characters in length. The entries in the rest of the lines represent the Cross Peak ID, ID’s of the correlated 1D ¹H peaks shown in parenthesis (for ambiguous correlations the IDs of all possible 1D ¹H peak correlations are included), peak intensity levels (which are classified as four types; strong (3), medium (2), weak (1), and unknown (0) - The default value is 3.), actual peak intensity or volume (optional, the default value is the peak intensity), and comments (optional, with a maximum size of 80 characters), for each NOESY cross peak, respectively. One or more space(s) is used as a delimiter for all items except comments which are separated by “;”. Items marked as optional can be omitted unless an item following it is included. In such a case, please include default values for ignored items even if they don’t get used. Comments can always be included as long as they follow a “;”. In this table, the numbers in the comment field correspond to the ID’s of the peaks in the SpecMan peaks table. For symmetric cross peaks, these numbers are shown with a + sign. For the NOESY peak list the peak intensity levels are used for calibrating NOE distance bounds in the NOE assignments.

II-10. Generation of Building Blocks

Once the SpecMan-generated peak lists have been converted, it is time to generate Building Blocks. During this step, NMR-SAMS interprets the molecular formula, ¹³C, ¹H, and HMQC spectral data (all data except ¹³C spectral data is optional), and generates possible building blocks for structure generation. To do this, select ‘Building Blocks’ from the Analysis menu and the following will appear:

The results of the building blocks process are written into the *.mdf file after the keyword ‘Atoms.’ Please note that for each step, NMR-SAMS keeps only one copy of the results in the master data file (*.mdf). When any step is repeated, the previous results of that step and any previous steps will be overwritten.

To modify the appearance of the building blocks, the user can select ‘Display Options’ with the following pull-right choices: Balls, Carbon Symbols, Numbers, Chemical Shifts, Protons, Molecular Formula, etc. from the Display menu.

II-11. Setting up Bond Constraints

To interpret all available 2D spectral data as bond constraints, select ‘Bond Constraints’ from the Analysis menu. In this step the various bond constraints, e.g. the H-H BC’s from COSY and the C-H ones from HMBC, are transformed into C-C bond constraints based on the HMQC-derived C-H connectivity. The following dialog box will appear:

NMR-SAMS warns the user that there are two very close ¹H peaks and that there is no COSY peak observed between them (suggesting that perhaps a near-diagonal peak between them has been neglected). Select ‘Yes to All’ to add ‘pseudo bond-constraints’ between these two ¹H peaks and for any similar ¹H pairs (16-17, 19-20, 21-22, 21-23, 22-23, and 26-27). The following ‘Summary of Constraints’ dialog box will appear:

Click ‘OK’ and the following building blocks with fixed bonds will be displayed:

Atoms with unsatisfied valences are marked by an asterisk (*), and are displayed in a different color (blue by default). Fixed bonds with unknown bond types are displayed as dashed lines, and can become single, double, or triple bonds in the structure generation process. Note that NMR-SAMS automatically sets up the carboxylic acid group based on the ¹³C chemical shift of C-1 and the available heteroatoms. For the Q-2 data, this is correct, but sometimes the automatically added functional groups may not be correct. In those cases, the user can modify them by selecting ‘User-Defined Bond Constraints’ from the Analysis menu.

NMR-SAMS also provides the user with the ability to interact with the atoms in the NMR-SAMS display window and the bond constraints listed in the Connection Table (shown below):

If the user clicks on a bond constraint in the Connection Table, the relevant atoms in the NMR-SAMS display window will be highlighted. Conversely, if the user clicks an atom in the NMR-SAMS display window, the relevant entries in the Connection Table will be highlighted.

II-12. User-Defined Bond Constraints

Once the bond constraints have been generated, it is important to review the building blocks, the fixed bonds, and the available bond constraints. For the Q-2 example, the generated building blocks, fixed bonds and bond constraints are correct. If a priori knowledge about the structure is known, the user can define additional constraints by selecting ‘User-defined Bond Constraints’ from the Analysis menu.

II-13. User-Defined Atom Environment Constraints

In addition to enabling the user to define bond constraints, NMR-SAMS also allows the user to define constraints on the neighboring patterns of an atom. For the Q-2 example, this step is not necessary, but if the user knows that C-4 must have one oxygen as its neighbor (based on its chemical shift 78.147), but does not know to which oxygen atom C-4 should be connected to, then it can be defined as an environment constraint of C-4. To do this, the user can select ‘User-Defined Atom Environment Constraints’ from the Analysis menu.

II-14. 2D Structure Generation

Once the user has given NMR-SAMS all the necessary information (plus any user-defined information) for structure generation, select ‘Generate 2D Structures’ from the Analysis menu. The structure generation process will be initiated, and the following dialog box will be displayed:

This dialog box shows the progress of the structure generation process. At any time the user can abort the structure generation process by clicking ‘Stop’ (it will take a few seconds to stop the structure generation).

In less than one minute, the following dialog box will appear to announce that one, unique, complete structure (and 49 partial structures) has been generated:

Click, ‘Yes’ to save the one complete structure and the 49 partial structures in ‘q-2test.str’ and the following dialog box displaying a summary of the structure generation results will appear:

Click ‘OK’ and the correct structure will be displayed as shown below:

The one, unique structure is displayed along with a structure browser so that the user can look through the 49 partial structures. The partial structures are listed in descending order of the number of generated bonds. The substructures provide useful clues when a complete unique structure is not generated.

II-15. Exporting NMR Data, Resonance Assignments and Structures.

NMR-SAMS enables the user to export chemical shift correlations. Select ‘Export – Chemical Shift Correlation’ from the File menu, and the chemical shift correlations will be written into a ‘q-2test.spc’ file.

To export the results of resonance assignment, select ‘Export – Assignment’ from the File menu, and the ¹³C and ¹H assignments of all the atoms in the selected structure will be written into a ‘q-2testx.rst,’ where x is the number of the structure. Note that the NOE cross peak assignments are appended to the end of the *.rst file.

To export structures into a structure file (*.mol, *.mdl or *.sdf), select ‘Export – Structures’ from the File menu. This will bring up the following dialog box:

Click ‘OK’ and the current structure will be saved in the ‘q-2test001.mol’ file. The structure elucidation process of Q-2 is now complete. The correct structure was obtained with complete resonance assignments.

II-16. Report Generation.

Now that the structure generation process has been completed, it is important to be able to create detailed reports. SpecMan and NMR-SAMS contain convenient clipboard functions for report generation. Re-initiate SpecMan, and select ‘Open Molecule’ from the Edit menu to open the ‘q-2test001.mol’ file (created in the step above) from the C:\Spectrum2001\Data\NMR-SAMS\Q-2\Q-2-MF folder, and the following molecule will be displayed:

SpecMan allows the user to renumber the molecule, since the user may want to use a different numbering system than that of the program. To renumber a molecule, select, ‘Molecule’ from the Edit menu. This will bring up the following Molecular Editor palette:

To renumber an atom, select ‘Renumber’ from the Molecular Editor dialog box. This feature will allow you to renumber atoms in a structure to conform to IUPAC or other conventions. The Molecular Editor dialog box will appear as shown below:

Select the number that you would like to begin renumbering with (for example, 1), and type it in the text box next to ‘Renumber.’ Then click on the atom in the molecule that you would like to renumber as atom 1 (for example, atom 13). Internally, atom 13 will now become atom 1, and the old atom 1 will be renumbered as atom 13. Additionally, the assignments will also be swapped so that they remain with the correct atom, and the assignment table will be updated with the new atom numbers. When you click on another atom, the program will increment the initial renumbering so that the next atoms will be labeled two, three and so on. If you had begun your renumbering with 20, then the next atoms would be labeled 21, 22, 23, etc. The renumbering of atom 13 to atom #1 is shown below:

Comparison of the initial molecule (shown on Page 55) with the above-modified molecule shows the old atom number 13 now renumbered as atom 1.

Once the molecule has been renumbered, select ‘Copy Molecule – Active Molecule’ from SpecMan’s Edit menu. Then, open a Word document side by side with SpecMan. Select ‘Paste’ from the Edit menu in the Word document to paste the molecule into the Word document:

If the user has opened up multiple molecules, all molecules in the molecular window can be copied and pasted into a Word document. To do so, select ‘Copy Molecule – Active Window’ from the Edit menu.

The user can also copy the assignment table, by selecting ‘Copy – Assignment Table’ from the Edit menu and then pasting the table into a Word document as shown below:

The assignment table will appear in text format in the word document. In order to convert the assignment table into actual table format, highlight the text portion of the assignment table, and under the Word document’s Table menu, select Convert Text to Table.

Similarly, the user can copy spectra and 1D and 2D Peaks Tables and paste them all into Word documents for easy report generation.

II-17. Resonance Assignment with NMR-SAMS

In some instances, the user already has some a priori knowledge about the structure, and is only interested in determining resonance assignments and not structure elucidation. Unlike other methods that obtain their assignments of ¹³C or ¹H chemical shifts solely from large databases, NMR-SAMS uses 2D NMR-derived connectivity information, in addition to standard chemical shift information. NMR-SAMS first uses standard chemical shifts to predict tentative assignments, and then utilizes 2D NMR-derived connectivity information to improve the tentative assignments to final resonance assignments. In this manner, the final assignments of NMR-SAMS are usually more reliable than the assignments based solely on predicted chemical shifts from large databases.

Target Structure-based Resonance Assignment

If the structure of Q-2 is known and the user is only interested in obtaining resonance assignments, the user should perform the exact same steps for structure generation in NMR-SAMS, but stop after the generation of bond constraints (section II-13). Then, select ‘Input Target Structure – Import Structure File’ and the following dialog box will appear:

Select the ‘q-2001.mol’ file and click ‘Open’ to display the known structure for Q-2:

Please note that the user can also choose to build the structure using NMR-SAMS' ‘Molecular Builder.’ To do so, select ‘Input Target Structure - Build Molecule’ from the Analysis menu. This will bring up the following ‘Molecular Builder’ palette:

Once the structure has either been built or imported into NMR-SAMS, select ‘Assign Spectra’ from the Analysis menu. The building blocks will automatically get mapped to the constituent heavy atoms in the target structure using all of the available constraints. In a matter of seconds, one complete assignment for Q-2 is obtained, and the following dialog box will appear:

The user is prompted to save the one complete assignment and the 16 partial assignments (useful when complete assignment is not possible). Click ‘Yes’ to save all assignments, and the information will be saved in a *.str file.

Then, the following dialog box giving a brief summary of the results of resonance assignment will appear:

Click ‘OK’ and the user can use the following ‘Structure Browser’ to look through the various assignments:

During resonance assignment NMR-SAMS tries to assign resonance to the best possible extent even if complete assignment is not possible. Resonance assignment starts from an internally selected atom, and if complete assignments are obtained during the first attempt, NMR-SAMS will stop searching all possibilities. On the other hand, if a complete assignment is not obtained during the first attempt, NMR-SAMS loops through different starting atoms and repeats the assignment process in order to get the most extensive possible partial assignments. By inspecting the unassigned portions of the structure, the user may be able to identify the regions in the target structure which conflict with the NMR data.

II-18. Structure Elucidation With Unknown Molecular Formula

NMR-SAMS enables the user who does not have a high resolution molecular formula, but has molecular weight information, to perform structure elucidation.

The procedure for structure elucidation with an unknown molecular formula is very similar to the procedure for a known molecular formula (utilize the q-2-nomf folder under Spectrum2001\Data\NMR-SAMS\Q-2). The differences are as follows:

1. When selecting ‘Create NMR Data File’ from the File menu, input the molecular formula as ‘unknown’ (see II-3).

2. During the peaks table conversions (see II-4 through II-9), the program will not check the results against a molecular formula, so some warning messages will not appear.

3. During generation of building blocks, NMR-SAMS will generate building blocks only for the observed ¹³C peaks, as shown below:

Thus the heteroatom building blocks will not be generated as in section II-10.

4. After the building blocks have been generated, the molecular weight (based on Carbon and Proton peaks) is displayed in the top left corner of the display window. The difference between this number and the actual molecular weight (in this case, 456.70) can be used as a guide to determine the elemental composition of the other possible heteroatoms (O, N, S, etc). In some cases, additional heteroatom information is also known from IR and UV spectral data. The user can use all available information to add heteroatoms by selecting ‘User-Defined Building Blocks’ from the Analysis menu. The following palette will be displayed:

In the ‘User-Defined Building Blocks’ palette, check ‘Add’, select ‘O’ for the ‘Element’, uncheck ‘Ignored Atom’, and enter ‘0’ for ‘Proton Count’ and ‘2’ for ‘Valence.’ Then, click in the NMR-SAMS display window to add one Oxygen atom at a time.

To add 2 OH groups, select ‘1’ for ‘Proton Count’ as shown below:

Then, click in the NMR-SAMS display window twice to add the 2 OH groups.

Once the building blocks have been modified, click, ‘OK’ and NMR-SAMS rearranges the building blocks and automatically sets up the ACMX again. The resulting building blocks should be the same as those described in II-10.

For the remainder of the structure elucidation process, the operations and results are exactly the same as those described for the example with a known molecular formula.

Computer-Assisted Structure Elucidation of Q-2 Using SpecMan and NMR-SAMS

Computer-Assisted Structure Elucidation of Q-2 Using SpecMan and NMR-SAMS

Separation (specifies the factor by which the base threshold is multiplied to determine the cutoff of each contour level)

Number of Levels (number of contour levels is by default 20 and can be modified in the *.ini files)

Starting Level (allows stepping through the available contour levels to resolve partially overlapped peaks and to locate peak tops)

Highest Peak

Computer-Assisted Structure Elucidation of Q-2
Using SpecMan and NMR-SAMS

Computer-Assisted Structure Elucidation of Q-2
Using SpecMan and NMR-SAMS