Cell Profiler assigning red channel to "Blue image" and DAPI to "Red Image"

swapped
wrong
splitting
channels

#1

Hello helpful people,

Since replacing the “AxioVision” software that controls our Zeiss scope with the newer “Zen” software, I’ve been encountering the following problem when trying to process my images using the same pipeline that previously worked:

The image that was collected from the blue channel, my DAPI staining, is being assigned to the “Red Channel” in CellProfiler. Similarly, the actual red image is being assigned to “blue channel”. This all happens in the ColorToGrey step. I could swap the names on my downstream processing i.e. calling my Red channel the DAPI channel, but that could get confusing and leaves me concerned as to what is going on to cause the issue in the first place.

I’ve uploaded a zip with my pipeline, an example folder of images that should just be dragged into the images file list and then filtered by clicking “apply filters to list”. There is also a screenshot of the result of the ColortoGrey step showing a faintly stained cell body in the “Blue Image” channel and DAPI staining in the “Red Channel”

Thanks in advance!

Help.zip (8.4 MB)


#2

I don’t know exactly what I’m supposed to be looking at, but are you sure you’re having the problem you think you’re having? This is a zoomed in crop of your Test_DAPI image you included in the folder:

These are crops of your screenshot’s blue and red images, respectively

And here’s the channel split done in CP with your pipeline on my computer and zoomed in on that particular cell.

All of that seems ok to me, so can you explain a bit more clearly what you expected to happen instead? Thanks.


#3

Thank you for your response. As you can see the blue image has a lot of staining that is not DAPI staining. This was a new issue in a staining that I have done hundreds of times. I’ve since had the Zeiss support staff out and its an issue with the way the new Zen software is exporting the .czi images as TIFs. The cellular staining in the DAPI channel is entirely absent in the original .CZI or in a TIF I generate via ImageJ.

Thank you for your help. This is not an issue with CellProfiler.