hello everyone,
I would like to ask how we can analyse CD45 cell (should red) by using imagej program automatic way but by accurate method ( i want the counting of the cell in field or the %)? do you think the threshold method could use or accurate ??? or there any way to do this analysis ???
Thank you for your help in advance.
Thanks
Nour Al Zaeed instead of counting, calculate the integrative density of the stain (this can be done by ImageJ). Then, count the number of DAPI dots for the same slice. After that normalize the results by dividing the integrative density of the satin on the number of DAPI dots. Do this for about 5 to 15 image then plot them on a graph and perform the desired comparison.
@Mohamed khashan
I am approaching your answer.
I thought there is another way of integrative density but if there on I will do.
Hi,
The first thing to do is to define a cell. To this end I strongly recommend you to use stardist in ImageJ to have a total count of cells.
https://imagej.net/plugins/stardist
This plugin yields ROIs corresponding to your nuclei, and this can be imposed over the CD45 image to measure intensities (Check the image attached, the segmented coloured images is the mask, and the last panel is the superposition)
Afterwards, you will have to set a threshold value (either in your image or your data) to define what is a positive (CD45+) cell.
By using the strategy I told you over the raw red CD45 image (after converting to 8bits because the original is RGB) i got the attached histogram (cells vs Fluorescence).
So the question is: Where do you set a threshold there?
Another strategy is to measure the fluorescence in both the red and the green channel and work over the ratios to try to define a threshold...
So the short answer is yes, the analysis is doable and can be automated, but you need a set a a threshold to differentiate CD45+ vs CD45- cells,
Hope this helps,
cheers,
J
To add to my previous answer,
If you are interested in a comparative analysis, assuming you have similar number of cells per field of view and your staining intensities are somehow similar between different experiments, you can try to plot directly the histograms and see whether there is any shift in the histograms in different conditions.
that way you don't need to set a threshold to define + vs - cell, but you'd rather focus on the distributions of the labeling intensities.
J. Ramirez-Franco and Mohamed Khashan Thanks for you to help me to do this analysis but Do you think that counting or integrative density in staining slides is an accurate method? maybe for cell ratio always FACS machine is better.
thanks so much for the help, without your answer I could not do it
I am so appreciate your help.
Nour
Nour Al Zaeed
Pls refer to the following article where they used the mentioned method:
Kim MS, Ahn JH, Mo JE, Song HY, Cheon D, Yoo SH, Choi HJ. Optimizing tissue clearing and imaging methods for human brain tissue. J Int Med Res. 2021 Mar;49(3):3000605211001729. doi: 10.1177/03000605211001729. PMID: 33771067; PMCID: PMC8166401
- Badges
- Science topic