I am trying to count cells that colocalize for two stains in a z-stack. I have got cellprofiler loading the z stack and counting, but I think its just counting all the 2d images separately and adding the numbers. I want to count the objects that overlap, in 3D. Any thoughts?
I took the images on an Olympus confocal, but I want to automate the cell counting
I think CellProfiler is not suitable for 3D image analysis. At the moment, CellProfiler can only analyse 2D images or individual images in a stack.
Most times the manufacturer of your microscope has some software for 3D visualization / rendering, or alternatively you could consider Imaris or Velocity. It's also possible in ImageJ etc, but this requires expert knowledge etc.
Do you know of any resources that could help me to get it working in ImageJ?
I would calculate the co-localization for all of the slices in a stack, and then identify a mean co-localization per cell by averaging all of the slices in the stack.
Ultimately, if your objects are truly co-localized then it should come up enough in each 2-dimensional stack to be representative when you average it. After all, if the two objects are only "co-localized" in separate slices of the stack, then maybe they aren't co-localized at all and are merely closely associated.
- Badges
- Science topic
- Similar topics
- Biological Science
- Molecular Cell Biology