Colocalization analysis

Hi,

I really hope i can use cellProfiler to get quantitative colocalization analysis… Can anyone help me to prepare the pipelines in order to get result…Im interested to know how many percent of my nanoparticles colocalize with lysosomes, endosomes and in cytoplasm…
really appreciate ur help!
:smile:
mc

What stains did you use and how did you capture the images?

Hi mchan,

If you are defining colocalization as “which small objects are contained within a larger object” (e.g., which cells/organelles contain particles), then you can use the RelateObjects to establish a parent-child relationship in which the smaller sub-objects are assigned as “children” to the enclosing “parent” object. This approach assumes that you are able to identify the sub-objects (nanoparticles) and larger objects (lysosomes, endosomes, cell/cytoplasm) with one of the Identify modules. From there, you can use CalculateMath or ClassifyObjects to calculate fractions of parents which contain child objects.

Hope this helps!
-Mark

Hi Feel,

I stained cells with draq5(nucleus), lysotracker and lysosensor(lysosomes) and FM464(endosomes)… I use confocal to capture the images…

regards,
mchan

Hi Mark,

Thanks for your suggestion… I will try and if having problem, I hope i can ask you for help…

tqvm,

regards,
mchan
:smile:

So do you need per-cell results or just a general resume of how much of your nanoparticle is located in endosomes, lysosomes and cytoplasm.

General)If you simply need to know how much of your nanoparticle stains coincides with endosomal,lysosomal and cytoplasmal stain. You can simply identify endosomes, then identify lysosomes and then nucleus.
Then you imagemath to sum all three and use it as and inverted imagemask to find how much is located outside this localitations.This would roughly equal cytoplasm if your nanoparticle stain is never located outside the cells.

Per-cell) If your stain also appears outside the cells then you would need to use the nucleus with identifyprimary and combine the endosomal-lysosomal stains to obtain a rough cell-limits with identify secondary and tertiary. From this tertiary you would then remove the endosomal and lysosomal locations obtaining a cytoplasm figure.

Determining a good object radius for id.primary from the begining will help you a lot.

At the end of the pipeline you add a colocalitation module to meassure how much of your nanoparticle stain coincides with each location.

This is a general idea, I know it may be explained better, but better handle a litle bit the program and then come when you find dificulties and we’ll try to give you a hand.

If you are scanning with the confocal monolayers or sections on which there is only one layer of cells then makeproyection will make life easier for you. If you have several layers you’ll have to meassure in each z-plane and then combine it at the database level.

People interested in this thread may want to take a look at our example pipeline and tutorial about colocalization here:
http://cellprofiler.org/examples.html#Colocalization