Measuring Degree of Clustering of Signal

Does anyone have any tips or thoughts on the best way of quantifying how clustered a signal is?

I have images of a fluorescently tagged protein in the cell that forms numerous punctate structures under normal circumstances. Upon induction with a drug, these they usually coalesce into one huge cluster in the cell. However, some of the mutants exhibit varying degrees of this ability to cluster. Ideally, I’d like to be able to generate some sort of way to detect and measure how clustered the signal gets. Here are some sample images:


Clustered and Some Clustering:

The green signal is what I’m interested in measuring. The white color in the second image is a dye for the cell membrane.

If anyone has any ideas, it’d be greatly appreciated, thanks!

Hello @neurofuzz,

If you are able to detect each punctuate structure (for example using Process > Find Maxima…), you can study their spatial distribution with the Spatial statistics plugin or the R spatstats package (that you can use from ImageJ as nicely explained by @Bio7 on this tutorial).


1 Like