This section provides protocols for MicrobeTracker use in order to perform a few most typical tasks. For a particular operation, the user may also need to read the description of the mentioned operation (button or tool) on their corresponding help pages.

See also available related tutorials that guide you through some of the mentioned operations on the example of provoded image sets:

• Cell detection (detection of E. coli cells in still images)
  
• Fluorescence profiles (detection of C. crescentus cells in time-lapses, obtaining fluorescence profiles)
  
• Accessing internal variables (obtaining a kymograph from a FRAP experiment).
• Extending MicrobeTracker (obtaining outlines and properties of nucleoids or other intracellular objects).

Usage for independent images

• Make sure you take phase contrast images, not DIC (DIC images can be converted into an acceptable format, however with a loss of quality). Place all phase contrast images into one folder. The tools renfiles and renfilestc can be used for this purpose: renfiles moves images from one folder into a set of subfolders, and renfilestc combines files from multiple folders and sorts them into new folders based on the name prefix. If you never used these tools for the particular file naming format, try first on a copy of the images.
• Load images clicking on Load phase data (and also Load signal 1 data and Load signal 2 data if necessary).
• Load parameters (usually alg4.set, your set, or the set from the previous working analysis), adjust necessary parameters (such as scaleFactor or areaMin, areaMax).
• Set the radio button on the Detection & analysis panel to Independent frames. Check if you want to save the results during processing.
• If you use fluorescence, subtract the background (check All and then click Subtract bgrnd for all signals).
• If you use fluorescence, check if you want to add signal(s) to the meshes (Compute signal 1 profile as / Compute signal 2 profile as).
• Try detecting cells on one frame to make sure the program works correctly. For this click This frame on the Detection & analysis panel.
• If the results look good, click All frames or put the beginning of the range to 2 (to skip frame 1) and then click on Range.
• When the program finishes (depending on the number of cells, if will take from a minute to several hours), check the performance. Delete occasional misdetected cells by selecting them and clicking Delete or pressing the keyboard Del button.
• Make sure you save the results (Save analysis) or they get saved to the right place during processing (select Save on each frame or Save when done, and chose the file name below).

You do not have to save the results during processing. However, this is preferable for large image sets in case the program crashes. In addition, there are memory leaks in MATLAB that cannot be easily eliminated, so that during long processing the memory requirements grow. If you are processing large image sets, it may be preferable to do the detection without fluorescence signal. In this case restart MATLAB after detection, load the phase contrast, the fluorescence images and the analysis, then subtract the background, check Reuse, and click All frames to add the signal without cells detection. In addition, if you have noticed excessive memory held by MATLAB in Windows task manager after running for a long time, restarting MATLAB may also be advisable.

Usage for timelapses

• Place the phase contrast images into one folder, signal 1 images (if any) into another, etc. See above for more details.
• Load images clicking on Load phase data (and also Load signal 1 data and Load signal 2 data if necessary).
• Load parameters (usually alg4.set, your set, or the set from the previous working analysis), adjust necessary parameters (such as scaleFactor or areaMin, areaMax).
• Set the radio button on the Detection & analysis panel to Timelapse. Check if you want to save the results during processing.
• If you use fluorescence, subtract the background (check All and then click Subtract bgrnd for all signals).
• If you use fluorescence, check if you want to add signal(s) to the meshes (Compute signal 1 profile as / Compute signal 2 profile as).
• If there is any noticeable drift (horizontal shift between frames as timelapse proceeds), align frames by clicking Align.
• Try detecting cells on a few frames to make sure the program works correctly. For this select frames from 1 to 3 (or from 1 to 2) on the Detection & analysis panel, then click Range.
• When the program finishes (depending on the number of cells, if will take from a minute to several hours), check the performance. Delete occasional misdetected cells by selecting them and clicking Delete or pressing the keyboard Del button. You can also remove them on one frame only and then with a separate script cleanCellList on the frames starting from the frame where the cell is missing. You can also redetect all the cells or a subset of them with a different parameters set starting from the frame where something went wrong.
• If the signal was not added simultaneously with cell detection, load the signal, subtract background, run in Reuse meshes regime.
• Make sure you save the results.

Usage for moving cells

If the cells are moving very little, the timelapse regime may still work. You can also increase moveall parameter to ~ 0.95. In most cases, however, you should follow the procedure below.

• Detect the cells using the independent images regime (see the description above).
• Remove misdetected cells from the first frame only.
• Save the results.
• Run trackstack command (read the command description for details). The simplest syntax is just typing trackstack, which will request the user to enter meshes file and the file to save the results.
• Load the results into MicrobeTracker and check the performance. If not good, try changing trackstack parameters.