What's the best way to measure the overlap between different color channels in cells (%red&green, %red&blue, etc.)?

Hi all,

I’m trying to do something pretty straightforward but I don’t know what the right tool is.

I have combined in situ and IHC of the spinal cord. I used 4 colors. I now want to determine the co-expression of these different markers.

DAPI = Blue
White = Gene 1
Red = Gene 2
Green = GFP transgene

I want to quantitate the co-occurrence of these signals, so for example (GFP+/Gene1+/Gene2-) etc in an efficient manner. I’ve used Photoshop in the past and used the selection too. But I know ImageJ must have a good way to do this.

I think it is best that I define the ROIs (what is a cell?) but then I pretty much want a way for ImageJ to look at all those ROIs and determine which of the colors co-occur in that ROI, so that I get a count of all the different possibilities. This is such a standard thing, but I’m trying to avoid doing it by hand/eye. When I have just two colors,I use photoshop, but there must be a better way.


Here’s an example of what I’m looking at.

1 Like

Look at coloc2 plugin in Fiji, or tools from mosaic update site. Decide if you want to do object based or intensity correlation based measurements depending on nature of the question. Sounds like object based to me.

I guess @chalkie666 was on the go, so here is a link to a wiki page describing the different procedures: http://imagej.net/Colocalization_Analysis