Help associating foci with nuclei in mouse embryonic retina cryosections

Hello all. I am looking for some pipeline optimisation suggestions that will allow me to count and associate nuclei and foci in mouse retina immunostaining images.

The attached pipeline initially splits the input image into two channels to give me dapi raw and Casz1 raw images. Then I use the dapi raw image to define an ROI that will be used for downstream analysis. What I want to do then is:

  1. Count the number of dapi nuclei within the ROI,
  2. Count the number of Casz1 foci within the ROI,
  3. Associate the number of dapi nuclei that contain Casz1 foci.

Basically I am having problems identifying the foci. I have changed several parameters in the ‘Identify Primary Objects’ module but I am still stuck with the same problems. I have attached my pipeline and a sample image if someone wants to try and help me out.

There is probably a simple solution for this but it currently escapes me!

Many thanks for any help,


Casz1Pipeline.cppipe (25.9 KB)

Hi Iwan,

I have seen your pipeline and its good that you are enchancing foci before detection. As I can see nuclei are so dense and it hard to separate from each other precisely. However you can try using RobustBackground thresholding method for enchanced foci and lower down further size of smooth filter and Suppress local maxima distance.

Good luck