Unexpected Results from MeasureObjectIntensity

Hello Fellow CellProfilers,

I have been using CP for a few months now to get some really great results in images we are trying to analyze, but recently I’ve noticed some weird results that I can’t explain and I’m hoping someone with more knowledge about the inside functions of CP could help.

I’m using CP to count spots (puncta) overlaying with DAPI. I’ve done all this with great success, but when I look at all the resulting data, there are times when an object is identified, categorized, and counted, but then yields zero intensity.

I’ve included pictures of an example of a cell I noticed today. The picture “4322_All_Expanded_Nuclei” shows the nuclei in question (the top-most orange cell). “4322_All_Gabra_Counted” shows the corrected brightness image and the selected puncta. “4322_All_CoGabra” shows the puncta spots that are within the nucleus pointed out above.

CP identified the nucleus, as well as 6 puncta spots (of this one channel) within the DAPI, which I think is correct. However, while it counted correctly that there were 6 spots, the integrated intensity was zero. The puncta in the other channel (no picture included, but can be if that would help answer the question) are significantly brighter and more filled out than in the picture I’ve included, however those puncta are counted correctly and their integrated intensity is never zero.

Does anyone know what might be causing this? Some options we’ve thought of are that the spots are very small so possibly when they are selected, there is not a region left in the middle that can be integrated, and thus becomes a 0. Another option is that there is some thresholding somewhere in the step of integration (that I didn’t set) and it ditches some of the data for some reason. A third possibility is that there is rounding of values somewhere along the way and that these are really dim spots get rounded to zero before exporting or by Excel later. For reference, one particularly dim cell was counted (correctly) with an integrated intensity of this channel of 2.6E-5.

If anyone knows the inner-workings of CellProfiler or Excel more and could help me understand why these objects are being selected but reported as 0 intensity, I would be very grateful.

Thank you all so much for your help and this great community!

Update: I found an error in what image I was feeding into the MeasureObjectIntensity module, which resulted in the zero intensity calculated. As far as I can tell, everything is working exactly as expected! Thanks anyways if you took the time to read this thread!