Processing multiple Images in CP for export to CPA


I have a question regarding CellProfiler Analyst – I have multiple images that I want to identify primary objects for in CellProfiler, and then export those objects to CPA. However, I know that CPA doesn’t like dealing with more than 1 set of primary objects. How do you get around this issue when dealing with multiple images? I know there must be a way because the demo pipeline handles hundreds of images. It would be very helpful if I could take a look at the original CellProfiler pipeline that was used to create the CPA files. I tried downloading what was posted here: … ?f=2&t=472

But it appears to be a .mat file, and I am not able to load it. Please let me know if you have any ideas.


Hello Vahid,

I think you misunderstand the object concept of CellProfiler. CellProfiler is cyclic in design, i.e. each pass though the pipeline objects are defined for each image set or field. Each image set will have its own unique objects yes, but all the objects from all the image sets are considered the same “object”. So when you classify your primary objects, say “Nuclei”, they will include the nuclei from image set 1, 2, …, etc.

Indeed if you have multiple primary objects, that breaks the data model for CPAnalyst and you either have to choose one object to classify, or be very careful in how you define and “link” your objects.

You might be helped if you go through the translocation exercise here: It takes you through an example CP analysis, data export to a database, and subsequently, CPAnalyst on the same data.