Hi everyone!
I am trying to calculate co-localization of two channels (red and green) in immunofluorescence tissue images. As read online, I am using the ImageJ plugin Coloc2 and performed some preliminary modifications (such as splitting channels, subtracted background and selecting ROI), but unfortunately I got always the same warning message as follows:
"The ratio of the y-intercept of the auto threshold regression line to the mean value of Channel 2 is high. This means the y-intercept is far from zero, implying a significant positive or negative zero offset in the image data intensities. Maybe you should use a ROI. Maybe do a background subtraction in both channels. Make sure you didn't clip off the low intensities to zero. This might not affect Pearson's correlation values very much, but might harm other results".
Does anybody of you have any hint/suggestion on how I can get rid off this warning message and obtain trustworthy results?
Thanks a lot in advance ;)
Francesca
This is hard to troubleshoot without example images. Can you post some?
Hi Francesca, I agree with Jordan (again),
I suppose you could also check a different co-localization plugin?
keep us notified
Hi,
Thanks a lot for your answers. I am sorry for the late answer, I was temporary off-line.
I post here some pics so you can have a clue of what I meant above in my question.
First is green channel, the other is in red.
I hope you might have some hints now on how I could solve my issue.
Thanks again!
Hi Francesca
the images you send are saturated in a lot of the signal and also have flat background. You should try to generate 12 or 16bit images with the setup you are using and then store them e.g. in tif format to keep the dynamic range of the original image.
You are getting the warning because the software recognizes that there are areas of maximum (and background) value. It cannot tell how bright a saturated pixel would be compared to the other pixels because when over exposed all pixels concerned have the same value. Same counts for the flat background were it assumes only an offset was added with no information. The background would usually be somewhat noisy so pixel intensity values would have a little variation.
If you are not familiar with the image acquisition software you are using you might get some local support there to optimize?
Hope that helps
Best
Heiko
Hi all,
I have the same problem as Francesca said at the beginning "The ratio of the y-intercept of the auto threshold regression line to the mean value of Channel 2 is high. This means the y-intercept is far from zero, implying a significant positive or negative zero offset in the image data intensities. Maybe you should use a ROI. Maybe do a background subtraction in both channels. Make sure you didn't clip off the low intensities to zero. This might not affect Pearson's correlation values very much, but might harm other results"
I generated 16-bit images, but I still don't know how to fix the problem. I appreciated any suggestions.
ps. I posted one of them
Thank you so much!
- Badges
- Science topic
- Similar topics
- Condensed Matter Physics
- Zero