Neuron Colocalization

Hello! I am new to CellProfiler and would appreciate some help.

I am trying to figure out a pipeline for counting colocalization using nuclei as the primary object/seed point. For some reason, when I try to identify secondary objects, the nuclei remain as objects when they are in a separate channel and I don’t see any major cross-channel interference.

My goal is to have a total cell count, total GFP-positive cell count (second channel), total PV-positive cell count (third channel), and total PV and GFP-positive cell counts. I am not sure what I am doing incorrectly here with my secondary object identification, and I would appreciate the help.

My (very short) pipeline and image I am trying to analyze can be found at this link.

I am using the most current version of CellProfiler, and have tried to base my pipeline on the instructions provided by David Logan on this forum.

Thank you!

Hi @golgitem,

I had look at your pipeline & images. The approach could be as follows,

  1. Segment the Nuclei (There is some filtering step required to remove the BG). This gives
  1. Segment the GFP channel I tried using Primary object itself & this gives
  1. Segment the PV channel I tried using Primary object itself & this gives
  1. Using “RelateObjects” module relate the GFP positive & PV positive objects (screenshot shown here) & this gives

Then if need measure the colocalization.
PFA pipeline
Enhancer Coloc Pipeline_LB.cpproj (919.9 KB)
Hope this helps.

Regards,

Lakshmi
www.wakoautomation.com

1 Like

@lakshmi Thank you so much for your help. I still have two questions-- first, I am noticing that a very clearly labeled GFP+ cell is not being picked up by the second identify primary objects module, but it is identified in the third identify primary objects module. How can I adjust settings to include this cell? I tried rescaling intensity for these channels without luck. (Edit: I figured this out! Was a diameter issue)

My second question is why RelateObjects doesn’t give a chart similar to IdentifyPrimaryObjects-- is cell count from the RelateObjects module just something I should add into my pipeline in a different way?

Thank you again for your assistance!

Hi @golgitem,

I think the first question you got the answer already.
Regarding the Relateobjects module, it doesn’t give the chart but you get the count in the results file. How this module works is this,
we are trying to relate a parent (gfp) to its children (pv). Every GFP will be related to pv if they are located specially close/same which results in the number of children’s count in the results file. In your case it will be only one child i.e. 1gfp object with 1 pv object (i.e. is renamed as GFP_PV).
For example, in the above screenshot (earlier reply) the # of GFP objects is 26 and pv objects is 16 and the third panel (below panel in the image) i.e. pv labeled by gfp is 7 (which your number of colocalised objects). This count you will get in the results file as GFP_PV count

Regards,

Lakshmi
www.wakoautomation.com