This project was established to help biologists detect late cytokinetic bridges on static immunofluorescence images. Icy is an open source image processing software developped at the Pasteur Institute. Within this framework, we provide this plugin and its code written to detect semi-automatically late cytokinetic bridges and evaluate their frequency in 2D cell culture, following immunolabeling and acquisition by fluorescence microsocopy. It requires the staining of nuclei, cytoplasms and midbodies.
Institution: Institut Pasteur
The cytokinesis plugin detects late cytokinetic bridges on static images. It uses Hierarchical K-Means to detect stained nuclei (DAPI, min intensity, min and max size in pixel) and midbodies (MKLP1, min intensity, min and max size in pixel) as well as a KMeans-adapted thresholder to delineate the cytoplasms (Tubulin, intensity classes). After expanding the ring containing each detected midbody (ring inflation in pixel), it interrogates the spatial relation between each inflated ring and the nuclei as well as cytoplasms. Only the rings without any intersection with nuclei and with at least two crossings of cytoplasmic areas are counted as part of late cytokinetic bridges. The left over midbodies are classified as remnants.
After having installed the plugin, please download the following archived images and files : Training_dataset. This archive contains a set of immunofluorescence images to test this plugin: HeLa cell; transfected with an siRNA anti-CEP55 or a non-targeting scrambled control; stained with anti-CEP55, -MKLP1 and -B-tubulin antibodies as well as DAPI. Drag the test images (.zvi) into the Icy window and run the plugin with the parameters suggested for this image set.