Kinetochore Analyzer - MontpellierRessourcesImagerie/imagej_macros_and_scripts GitHub Wiki
Segment kinetochores in yeast cells and measure the co-occurrence with a molecule in a different fluorescent channel. Two measurement approaches are available: the classic overlap-based approach and a dilated approach.
- You must have cellpose 3 installed.
- You must have the MorphoLibJ-plugin [1] installed.
- You must have the PTBIOP-plugin installed.
- Download the cellpose model and the labkit classifier and note the paths under which you saved them.
Download the file kinetochore_analyzer.ijm and put it into the macros/toolsets-folder of your FIJI installation. Restart FIJI and select the kinetochore_analyzer-toolset from the switch toolsets-button (>>).
Press the install or update button and follow the instructions. This installs the scripts of both measurement approaches into plugins/kinetochore_analyzer/. The dilated approach is available as the menu commands Analyze Dilated Kinetochores and Batch Analyze Dilated Kinetochores in the Plugins menu.
A right-click on a tool-button opens its options-dialog.
- signal channel
- The channel containing the signal that will be measured.
- kinetochore channel
- The channel containing the kinetochores.
- cell channel
- The channel containing the cells (can be the same as the signal channel).
- cell diameter
- The diameter of the cells in the image in pixel to match to the diameter of cells with which cellpose was trained.
- conda env path
- The path to the conda environment containing your cellpose installation.
- model path
- The path to the cellpose model.
- labkit classifier path
- The path to the labkit classifier used to segment the kinetochores and the signal.
- spot label
- The label representing the kinetochores or the signal in the results of the labkit classifier. If you are using the provided classifier the value is 2.
- background subtraction value
- The value which is subtracted from the signal in each cell before the signal is measured. It can be the mean, the maximum, the mode or the median of the signal intensities in the cell. The value is computed on the cell region excluding the kinetochores and the signal regions, and is then subtracted from all signal pixels of that cell.
- use gpu
- If selected the cellpose model is run using the gpu. You need to have an appropriate gpu and corresponding cuda version installed. If not selected, the model is run on the cpu.
- image file extension
- The file extension of the files you want to analyze.
The dialog of the dilated approach contains two additional options:
- kinetochore dilation radius
- The radius in pixels by which the kinetochores are dilated before the signal is measured. The dilated regions are restricted to the cells.
- signal threshold
- A threshold in intensity applied to the background-corrected signal. Only pixels with an intensity greater than or equal to the threshold are measured.
The cells are segmented with cellpose 3. The kinetochores and the spots in the signal image are segmented with a labkit classifier. Kinetochores and signal outside of cells are removed as false detections. The kinetochores are labelled with the labels of the corresponding cells.
For each cell, a background value is computed from the signal intensities in the cell region excluding the kinetochores and the signal regions. Depending on the option background subtraction value either the mean, the maximum, the mode or the median of these intensities is used. This value is subtracted from all signal pixels of the corresponding cell, resulting in a background-corrected signal.
Overlap approach (Kinetochore Measurements): The kinetochores and the signal are segmented independently, both restricted to the cells. For each cell with kinetochores, the background-corrected signal is measured over the complete segmented signal regions of that cell, i.e. not only over the part of the signal that overlaps the kinetochores. This was chosen because the signal spots sometimes lie next to the kinetochores rather than inside them. No dilation and no signal threshold are applied. Cells with kinetochores but without any signal are reported with zeros.
Dilated approach (Dilated Kinetochore Measurements): The kinetochores are dilated by the kinetochore dilation radius and the dilated regions are restricted to the cells. Only pixels of the background-corrected signal with an intensity greater than or equal to the signal threshold are measured. One entry is created per cell that has kinetochores. Cells with kinetochores but without a signal pixel above the threshold are reported with zeros.
The control image shows a maximum projection of the input image in which the signal channel is replaced by the background-corrected signal, so that the intensities displayed and overlaid with the measured regions correspond exactly to the values that are measured. The cell contours and the kinetochores are shown as overlays, and the measured signal regions (or measured zones) are highlighted.
The area, mean intensity, integrated density and the raw integrated density are reported. As in FIJI/ImageJ, the integrated density is the mean intensity multiplied by the area of the object in physical units. The raw integrated density is the sum of all pixel values or the mean intensity multiplied by the number of pixels of the object.
- Legland, D., Arganda-Carreras, I., & Andrey, P. (2016). MorphoLibJ: integrated library and plugins for mathematical morphology with ImageJ. Bioinformatics, btw413. doi:10.1093/bioinformatics/btw413
Volker Bäcker