Hello. I am analyzing z-stacks collected from layers of mucus secreting cells in whole mounts of gill epidermis (no mechanical sectioning involved). The data are collected with 2-P microscopy, in 3-D stacks.
a) The mucous cells are densely packed. The mucus granules within each cell are labeled with a fluorescent conjugate (SBA- Alexa 594: “purple” channel in image).
b) The plasma membrane of each mucous cell is labeled with fixed FM 1-43- FX, in the “green” image.
Task: I need to measure the mean/ median SBA- Alexa 594 values, as well as the mean/median mucus granule size for each cell. I’d like to identify the boundary of each cell based on the FM 143 plasma membrane labeling.
There is little/ zero pixel overlap between the Alexa 594 and FM 1-43 labels. I do not want to measure labelling colocalization at the pixel level.
I can do this in Volocity, which allows me to map each “blob” of mucus to the surrounding correlated FM 143 labelled membrane. Using Volocity, I was able to identify each plasma membrane as an object and the mucus “blobs” as objects within objects. Volocity reported the size/ intensity values for the mucus blobs. However, I have to pay to use Volocity.
I would like to do something similar with FIJI, but it seems most or all “colocalization” pluggins are aimed at treating pixels as objects (for molecular colocalization analysis on a pixel to pixel basis) and not for measuring correlation/colocalization between cellular structures such as nuclei, E.R. and plasma membranes.
I uploaded some images in a montage, and individually w/ .png format. The images are flattened sub-stacks of a 3-D stack. I hope this is clear (I’m a new user).
<a class=“attachment” href="/uploads/imagej/original/3X/5/6