Seriously amature problem - the signal inside nuclei



Hi everyone, I’m relatively new to Cell Profiler and I’ve run into a simple problem. I want to quantitate the signal from my red channel only inside of my nuclei (blue channel), but I don’t care about identifying individual nuclei (also, the “identify primary automatic” module doesn’t work too well on the DAPI pictures I have). What I’d like to do is use something like the binary image that “Measure Image Area Occupied” module produces, which outlines my DAPI really well and use that as a mask to exclude any part of the red image that isn’t inside it and then just find the total integrated intensity of all the red inside the DAPI. Seems like it should be easy to do, but I can’t seem to figure it out. Any help is greatly appreciated!


The method below should work, if you want the mean intensity across all nuclei in an image:
(1) MeasureImageAreaOccupied on DAPI channel producing image called e.g. “CellStain”
(2) Crop module on DAPI image

  • with “Into which shape…” set to “Other…” -> “CellStain”
  • producing e.g. “CroppedCellStain”
    (3) MeasureImageIntensity on “CroppedCellStain”
    (4) ExportToExcel module, choosing “Image” data
  • Look at “Intensity_MeanIntensity_CroppedCellStain” column of output Excel file

A side note: IdentifyPrimAutomatic and MeasureImageAreaOccupied utilize the same thresholding algorithms, and if you turn off all the other settings in IDPrimAuto, you should be able to essentially duplicate the segmentation. So I’m not sure why your IDPrimAuto doesn’t work well. But if you don’t need the specificity of IDPrimAuto settings, or per-object measurements, then using the above is fine.



Thanks for the advice, David. I think that will work well, but I’m not quite sure how to save the image that Measure Image Area Occupied module creates. There’s no option to name the image in the module itself. Do I need a newer version of Cell Profiler? Thanks!


From my understanding, if any modules end up created objects, to which you give it a name, then you can use convertoImage and Save images in combination to save your module-identified objects.

If you can’t seem to convert to image, then you’ll need to save the whole figure, .fig - I think CP help says you need matlab to open the file, but I found that if you point your input folder to the folder where your .fig are located, you can open them up by double clicking in the bottom left box.



Sorry, I should clarify - there’s an option that says “What do you want to call the staining measured by this module?” and if I input “CellStain” there and then try to load “CellStain” later with the Crop module it says there is no image by that name. I’ve also tried the “SaveImage” module but even when I input the module number to save the whole figure i(i.e. “5”) t tells me that there’s no image by that name. Am I doing something wrong?


Hi guys, I just downloaded the newest version of CellProfiler and now David’s advice works perfectly. Thanks for the help!