Count double and triple labeled cells

I downloaded CellProfiler yesterday and I want to make certain that I am doing things correctly. I want to quantify the number of triply stained neurons I have grown in culture. They are stained with DAPI, NeuN as a cell-type marker, and cleaved Caspase3. As one of my stains is a little dirty, I want to make sure that any NeuN or Capase objects map to a DAPI object. My goal is to come the grid below.

          | caspase - | caspase + |
----------------------------------------------
   NeuN - |     A     |     B     |  A+B=E
----------------------------------------------
   NeuN + |     C     |     D     |  C+D=F
----------------------------------------------
          |   A+C=G   |   B+D=H   | E+F=G+H=I

I = all DAPI positive cells 
F = NeuN+DAPI+ cells
H = Capase+DAPI+ cells
D = NeuN+Caspase+DAPI+ cells

The above is what I am trying to measure, and everything else can be easily calculated. I think I am doing this correctly, but want to make certain I am not missing something. Any suggestions?

Thanks in advance.
04082015C-pipeline.cppipe (34.5 KB)





I think you have much the right idea. I’m attached a revised pipeline, in which I did the following:

  • I replaced RelateObjects with MaskObjects, as RelateObjects enforces that an overlapping child object can only be assigned to one parent, whereas MaskObjects can assign the same overlapping object to multiple parents.
  • The FilterObjects strategy is otherwise the same as yours.

As long as you are assured that the NeuuN and Capase identification is sufficient, the basic approach should work fine.

Regards,
-Mark
assay.cppipe (23.4 KB)