Relate DAB staining to nuclei


I’m trying to use Cellprofiler to quantify DAB staining in tumor tissue. My protein of interest are nuclear and I try to find a way to measure the intensity to compare samples before and after treatment with a drug. Ideally, at the end I can get the percentage of positive cells.

I already started to modify a pipeline that is given in the forum but I’m not satisfied with the results …in particular the outlined measured size of DAB stained parts are smaller than the corresponding nuclei. I tried to figure out a parent-child relationship without succes :frowning:

Thank you for the help

pipeline_DAB.cppipe (12.2 KB)


You seem pretty close! Please see my attached pipeline which I think helps.

Here are some notes:

  • I turned off the Metadata module, since it wasn’t working for the image provided anyway
  • IdentifyPrimaryObjects: The main changes were here, especially in the “clumped objects” settings. I turned off the automatic settings, since the cells were being broken up too much and set the “smoothing filter” and “local maxima” suppression sizes manually. I also changes the method to distinguish clumped objects to Shape (vs. Intensity) since intensity is almost saturated and the cells are more-or-less round so Shape should work better (imho).
  • I didn’t change this, but if you want per-cell measures, you should change MeasureImageIntensity to MeasureObjectIntensity. But perhaps you don’t care and only need per-image measures which is fine.
  • ExportToSpreadsheet now has a red triangle so make sure you are outputting the measures you care about

Hope that helps!
DL_DAB_pipeline.cppipe (12 KB)