Question about Fluorescence quantification with IntDen

Hello everyone,

I am new in this forum and I was wondering if you could help me a little bit.

My aim is to quantify the fluorescent signal (nuclear and cytoplasmic) in treated and untreated samples.
After subtracting background and setting the threshold, I tried 3 different ways:

  1. I just did ‘Measure’ and I got for example:
    Area= 1240.45; Mean=13.494; IntDen=16738.21
  2. I did Analize Particles (size and circularity=0 for this example) and I got:
    Area= 1216.175; Mean=6.552; IntDen=69.696
  3. I applied the Threshold and converted to a mask and I got:
    Area= 1216.175; Mean=254.925; IntDen=1318.851

Which one would be the best approach and, I know that the IntDen= Area*Mean, but why the IntDen is so different in the 3 cases?

Looking forward to your reply.

Thanks a lot!
Best regards,
Tommy

The mask is likely 0 or 255. You got the mean value of a bunch of 255 pixels, so the 3rd case is area*255. It really doesn’t have much of anything to do with your image, other than the area measurement maybe being accurate.

Not sure about what exactly you did in the top two.

what I am aiming at is to quantify the fluorescence signal and compare treated and untreated samples.
After having selected different ROIs of the same size, I subtracted the background and set the threshold.
After that, in the first case I used the ‘Measure’ command to measure Area, Mean and IntDen limited to the threshold (checked ‘limit to the threshold’ option in ‘set measurements’).
In the second case, after setting the threshold, I used Analize Particles to calculate Area, Mean and IntDen and my question is: which is the best way to quantify the intensity of a ROI, using the Measure command (limiting to the Threshold) or using analyse Particles?

Thank you
Tommy

I’m afraid the description could mean too many things, and different types of “background” could mean different things or be treated differently. How you are determining your ROIs can have a major impact on that as well. One cell? One well? Tissue area? There isn’t really a one size fits all answer.

Thanks for your answer.
Briefly, I have done immunofluorescence staining (488, in Ch2) of paraffin sections and I am interested in tissue area, actually in the total signal in a specific tissue area that is localized in the same position in all the samples. What is commonly done in my specific case is to subtract Ch3 (autofluorescence) to Ch2 and, in this way, I have only the true signal.

Now, my question is simply what is the best way to quantify the total fluorescent signal in a specific tissue area? Just usign the ‘Measure’ command or using the ‘Analyse particles’ command?

Thank you

1 Like

Still not sure, but have you looked at this thread and the links about IntDen?

And hopefully someone else might come along with a more definitive answer.

1 Like

Yes I did but I haven’t found and answer yet. What I’m looking is just at what is the best way to quantify the total fluoresnce signal in a tissue section…

You should not threshold if you are intending to compute a measure of the grayscale intensity in your objects. So the 3rd measurement is meaningless as far as fluorescence quantification of the original image is concerned.
Check the calibration of the image, maybe the difference in the two figures comes from that.
Look for “RawIntDen” in this page
https://imagej.nih.gov/ij/docs/guide/146-30.html

thanks for your reply!
Yes, indeed I could use the RawIntDen since it gives the total number of pixel values.

For the threshold, basically I don’t ‘Apply’ the threshold (that converts it in a mask of 0 or 255 values) but I just do ‘Set’ threshold. In this way each pixel keeps its original value but it allows you to quantify only the thresholded pixel (if you set Limit to threshold in Set measurement).
…at least this is my naive understanding of it…

For the calibration, I would not know how to deal with that because I didn’t use any physical beads or method to create a calibrated curve. However, I took all the images at the same time and with the same confocal settings.

That’s the first time I have to quantify fluorescent signal in tissue sections and that’s why I wanted to hear different opinions about that.

Thank you