How to analyze punctuate fluorescence in multiple selections?

Image Analysis
, ,
#1

Hello everyone,

So I have to analyze punctuate fluorescence (see example image in red) in cells that expressed EGFP fluorescence (see example EGFP image).
My images are from a Z-stack obtained by confocal microscopy. In my initial image, I split the green and red channels, and I do a Z-project for each channel (Max intensity) followed by substraction of background in red and green channels.

Then, I threshold the red channel followed by “analyze particles” tool, and it works pretty well.
The problem I have is that I only want to analyze particles that are in a given overlay (EGFP cell contour I draw in the green channel). I didn’t find how to restrict analysis of red fluorescence in all of my contoured EGFP cells.
For the moment, I contour individually each EGFP cell, then Ctrl-MajE in the red channel, and finally analyze the particles in the given contour using the ROI I created with the thresholded image. As you can imagin, this manipulation cell by cell is very time consuming and I wanted to automatize the analysis to measure the particles in all my contoured EGFP cells at once.

Sorry if my english is not good, I hope I has been understandable.

Thanks a lot,

Analysis goals

EGFP example
EGFP example1024×1024 115 KB
red example
red example1024×1024 97 KB

#2

If you have a method to segment/threshold your cells within FIJI, you can create a Mask during the Analyze Particles… step.

If you adjust that Mask so that it is 0 or 1 (not 0 or 255, maybe by dividing the Mask by 255), you could then multiply the mask by the red image. Everything that is outside of the cells will now be 0 - multiplied by 0, and everything inside of a cell will have it’s same intensity and can be analyzed as normal.

Roughly -
Make a mask of the green (however you threshold it)
Adjust the Mask values to 0 or 1 (Image math)
Image Math, multiply red image by mask
Analyze resulting image for red spots.

#3

Thank you ! I will try with this method, hope it will works well

1 Like
#4

Hello,

Thank you, because it’s really simple to do and it works really well.
I just have a last question : Do you know if it’s possible to automatically attribute each counted particle in a given contoured cell ? It’s actually useful for my future analysis, for example, how many particles, or their fluorescence intensity I obtained depending on the size and/or green fluorescence intensity in each cell.

Thanks again,

FJ

1 Like
#5

Afraid that might be a topic for a new… topic. I typically use QuPath for more complex analysis due to it’s hierarchical structure, which makes outputting what is inside of each object very easy - for me at least. And the drawing tools are very flexible and powerful.

One note, if you are manually drawing cells, you may want to look into the wand variant created by @jerome

That is the closest I have seen to QuPath’s annotation tools, although the “prevents overlap” option in QuPath is nice.