Web based two-dimensional NMR spectra viewer

1. This is a web-based tool to view and process 2D NMR spectra. This tool is part of the frontend of the more advanced COLMAR deep picker server.

2. Upload one or more 2D NMR spectra in nmrPipe .ft2, Sparky .ucsf, or Topspin .txt (using command totxt) format using the "Choose File" button above. Additional files can be added by repeating this process. You can also drag and drop files into the dashed area titled "Uploaded Spectra". If you drop a folder, all supported files within the folder will be processed sequentially.

3. The server will automatically estimate the noise level of each uploaded spectrum. By default, the lowest contour level is set at 5.5 times the noise level, and each subsequent level is 1.5 times the previous level until reaching the spectral maximum. Users can adjust both values and recalculate. Users can also click on the down button to add one more contour level (current lowest level/ Logarithmic scale) at the beginning of all levels. This process is more efficient because there is no need to recalculate all other levels. The slider can be used to set the lowest visible contour level without recalculation. Users can also change the color of each spectrum.

4. Users can brush an area in the main plot to zoom in and use buttons to zoom out, zoom back, or reset all zooms.

5. Users can drag and drop spectra to re-order them on the plot.

6. Users can show horizontal and vertical cross sections of the last spectrum in the spectral list.

1. For each experiment, click on "DEEP Picker" to run peak picking. The current lowest contour level will be used as the minimal peak height cutoff.

2. After Deep Picker finishes, you can edit the picked peaks by (1) dragging them to the correct position or removing them by dragging them to an empty spectral region; (2) removing all peaks in an area by brushing; (3) adding a new peak by clicking on the plot.

3. Click on the button to run either Voigt profile or Gaussian profile peak fitting.

4. After fitting, a reconstructed spectrum will be added to the plot, which can be removed. The picked and fitted peak list will also be associated with this new spectrum.

5. You can rerun DEEP Picker and Peak Fitting, using new parameters. Each new peak fitting will create a new reconstructed spectrum.

6. You can download the picked, fitted peaks list and show or hide them by clicking the buttons and checkboxes.

1. Upload the first plane of your pseudo-3D experiment.

2. Adjust the lowest contour level, then run Deep Picker; only peaks above the lowest contour will be picked.

3. Edit peaks if necessary.

4. Run Voigt_fitter to optimize peaks. Repeat steps 2-4 if necessary.

5. Upload the remaining planes of your pseudo-3D experiment and ensure the fitted peaks of plane 1 are visible.

6. Run pseudo-3D fitting.

7. Optionally, you can upload an assignment file (Sparky .list), which will be transferred to the fitted peaks of pseudo-3D fitting.

8. Download the results from pseudo-3D fitting.

This program utilizes WebWorker and WebAssembly, which can't be loaded automatically when running locally unless you add the required command line to Google Chrome. To do so, right-click on the Google Chrome icon, select "Properties", and add "--allow-file-access-from-files" to the end of the "Target" field to look like this: "C:\Program Files\Google\Chrome\Application\chrome.exe" --allow-file-access-from-files then click "Apply" or "OK". After this, click the Google Chrome icon to run the browser first before loading. Unfortunately, adding this option poses a security risk. Therefore, do NOT load any local files unless you are sure they are safe.

1. Currently only supports 2D or pseudo-3D. For NUS, only supports 2D.

2. Upload the fid file (usually called fid or ser) and acquisition file (usually called acqus) for the direct dimension.

3. Upload the acquisition file (usually called acqu2s, or acqu3s for pseudo-3D) for the indirect dimension.

4. For NUS experiment, upload the NUS list file (usually called nuslist). Indirect phase correction must be provided.

5. You can also drag and drop the Bruker folder (which contains the required files) into the dashed area.

6. Set the apodization function, zero filling factor, phase correction for both dimensions. Default values work fine for most cases

7. Click "Upload experimental files and process" to process the spectrum.

8. If you want to reprocess, click the "Reprocess" button in the spectra list. The selected spectrum will be highlighted. Do you reprocessing and click on Quit reprocessing after done.

1. After processing the spectrum, you can manually adjust the phase correction for both dimensions by first enable reprocess.

2. Select "Cross section" (instead of "projection") to show the 1D cross section of the spectrum.

3. Move your mouse to either direction or indirect dimension cross section plot. Hold shift key down then use your mouse wheel to change P0. Each wheel event will change P0 by 1.0 degree. You can also use the Ctrl key for fine tuning (0.2 degree per event).

4. Click on the cross section plot to set anchor point. Right click to unset anchor point.

5. When anchor is set, use mouse wheel to change P1.

6. After you are satisfied with the phase correction, click "Apply PS" to apply the phase correction to the 2D spectrum and saved processing parameters.

1. Background peak picking/fitting gets stuck.

WebAssembly has a hard memory limit of 4GB. If your spectrum is very large with lots of peaks, Deep Picker or Voigt Fitter may get stuck due to running out of memory. You can try reducing the size (ZF) or increasing the lowest contour level.

2. Pseudo-3D fitting takes forever to finish.

The pseudo-3D fitting is a relatively expensive operation, requiring a lot of computation power. Because WebAssembly is single-threaded and has limited memory, it is not suitable for large pseudo-3D fittings. If you have a large pseudo-3D experiment, you can still use COLMAR Viewer to optimize the initial peak list from the 1st plane and then run pseudo-3D fitting using the standalone Voigt Fitter program.