Announcement of new ImageJ plugin: A versatile toolbox for semi-automatic cell-by-cell object-based colocalization analysis

Link to paper:

Video explanation (part 1):
Video explanation (part 2):

FIJI update site (add to install plugins): Index of /ObjectColocalizationPlugins


ImageJ plugin 1: Colocalization Image Creator:
*Pre-process multichannel Z-stack (or 2D) microscopy images into a visual format for faster, simpler, and more accurate colocalization analysis.
*Designed to help avoid common colocalization analysis artifacts and errors.
*Can transform Z-stack 3D data into a specialized 2D Z-projection where Z-projection colocalization artifacts are removed/reduced. This simplifies the analysis of 3D colocalization data.

ImageJ plugin 2: Colocalization Object Counter:
*Quantity (count) cells/objects in a semi-automatic manner.
*Assign, classify and keep track of multichannel signal presence/absence (colocalization analysis) for each cell/object.
*Tools for subsequent 3D modeling/representation of data: draw tissue contours and indicate image-series global XY-origin.
*Save data, load data, and export data to Excel.

Custom Excel macro-file:
*Import data from Colocalization Object Counter
*Analyze and edit data from image series.
*Export combined image series data to Matlab for 3D modeling

Custom Matlab script:
*3D visualize cells according to colocalization data
*3D visualize tissue contours

I hope the community will appreciate our work. The ImageJ plugin 1 might be somewhat hard to understand how to use effectively (though we hope not), but ImageJ plugin 2 should be very simple and useful to the broader community. I found a good cell counting tool for ImageJ lacking, so maybe this plugin (and the other) can be included as a standard part of FIJI.

Anders Lunde, PhD
University of Oslo, Norway


Thanks! Is this mainly for thin 3D data (tissue slices from cryosections etc) or also for bigger pieces of 3D data (cleared tissue etc)?

It can be used with any type of 3d data. But the custom Z-projection works best when cells are not excessively stacked in the Z-dimension (multiple cell layers), and that of course depends on the density of target cells and the Z-thickness of your data. However, if your 3D data is too thick (and dense) along the Z-dimension, you can always reduce your stack into smaller lengths using the ImageJ Image > Stacks > Tools > Make Substack command.

Ok thanks :slight_smile: Very cool and helpful plugin by the way!