iCRAQ

ImageJ macro developed to analyse chromocenters in plant nucleus¹

This macros rely on ActionBar, ImageScience and SCF MPI CBG plugins.

To run iCRAQ place the iCRAQ folder containing the iCRAQ_main.txt file inside the macros folder of ImageJ, launch ImageJ then go into Plugins>>Macros>>Run... and choose the iCRAQ_main.txt file:

• drag and drop one (or more) .tif (or .lif) files of microscopy z-stack on the iCRAQ buttons menu, the macro will open the first file (or the first serie of the first .lif file)
• Tune the options via the Option button if needed (see the options.txt for more details)
• Proceed with the nucleus detection by clicking on Detect Nucleus:

• Missed nuclei can be added (one at a time) by clicking on the Manualy Add Nucleus button and then using the freehand tool (or any other ROI drawing tools)
• Unwanted nuclei ROIs can also be removed (one at a time) with the Remove Nucleus button and then choosing the ROI of the unwanted nucleus in the Roi manager
• Fuse Nucleus button is mainly used to fuse elongated nuclei that have been cut by the nucleus detection process or to group nuclei to be removed:

• Proceed with the chromocenter detection hitting the Detect Chromocenter button
• Each detected nuclei will be analysed separatedly
• A first dialog panel allows to skip the chromocenter detection of the current nucleus by answering Yes:

• Otherwise you will be asked to adjust the sliders in the H-watershed plugin to get a good segmentation of the chromocenters:²³

• Then export the chromocenters mask via the H-watershed plugin menu and click Ok on the pop-up window to proceed to the next nucleus⁴

• As for nuclei, missed chromocenter can be added manualy with the Manualy add Chromocenter button and first selecting the nucleus ROI from which the missed chromocenter belong to and then drawing it manualy
• Chromocenters can also be removed with the Remove Chromocenter button and clicking the ROI of the unwanted chromocenter.
• Quality of the segmentation can be shown on the 2D projection via the Show On Projection button before performing the quantification (close the projection before running the quantification):

• The segmentation can then be saved using Save annotations which will first ask the name of the annotation image, then the destination folder and will save a .tif image with 2 or 3 gray levels:⁵

• Finaly, the Analyse Selections (2D) will ask for a result file name, perform the quantification of different ROI parameters and save the informations in 2 separate .txt files one with nuclei informations and another one with each chromocenter informations⁶

• Note that the data from different images corresponding to the same serie can be appended in one unique file by providing the same result file name and not checking the Delete if exit option

• The Next Serie button will close the current image and open the next one in the queue
• The Reset button remove all ROI and reset the number of nuclei detected to 0

Footnotes

1. This macro was developed during an internship in IBENS under the supervision of Fredy Barneche and Clara Bourbousse. ↩

2. Tips: change the display style to Contour overlay, uncheck the Allow Splitting box, tune the Seed dynamics to have one watershed region per chromocenter, adjust the Intensity threshold to fit to the chromocenter and finaly decrease the peak flooding if needed ↩

3. The View image in the H-watershed plugin menu allow to choose between the chromocenter features image (corresponding to the projection of the first eigenvalue of the structure tensor) or the projection of the original stack ↩

4. Note that the H-watershed plugin run in a java window that cannot be closed easily via ImageJ macro language, and the plugin windows will accumulate. In addition on each nucleus crop the ROI of the previous nuclei get overlaid, this is a known display issue but doesn't impair the quantification. ↩

5. The gray levels will be: 0 for background, 128 for nuclei and 255 for chromocenter if chromocenter are present in the image, or 0 for the background and 255 for the nuclei if no chromocenter are prensent. ↩

6. The measured ROI parameters includes: the x, y position in the image, the area occupied, the mean and standard deviationof gray level, the raw and normalized sum intensity and shapes descriptor (roundness, circularity, aspect ratio and solidity). For the nuclei the number of chromocenter (#CCs), the Relative Heterochromatin Fraction (RHF, the fraction of the nuclei signal intensity inside the chomocenter), the Relative Area Fraction (RAF, the area of the nucleus occupied by all chromocenter) and the chromocenter total intensity and area (CCintden and CCarea respectively) are also measured. ↩