Identifying "halos" around objects


I am using CP to automate counting of bacteria within autophagic vacuoles. About 10-15% of the bugs are within these compartments, which we mark with a fluorescent reporter. I can successfully identify cell nuclei, bugs and the vacuoles, but cannot work out how to get a count of bugs within vacuoles.
I have nuclei, bugs and autophagy-reporter-bright objects all ID’ed as primary objects and the segmentation appears robust. I though that IDsecondary object would work, but I guess it assumes that all objects (bugs in this case) have an outer secondary object, whilst in my case only 15% do.

What is the simplest way to implement this? Can I find bug objects within or overlapping the autophagosome object?

Will CalculateImageOverlap work - it seems to be written for testing/comparison purposes more that this application?



Hi Alan,

This sounds like a job for the RelateObjects module. It establishes a relationship between objects (“parents”) and sub-objects (“children”); in this case, the children are the bacteria and the parents are the autophagosomes. You can then get a count of bacteria per autophagosome object plus the mean of whatever bacteria measurements you’ve already made.


Thanks, I got the relate function working and it is indeed what I needed.