Synergisic between image cytometry (using FCS Express) and multiplexed image analysis using QuPath

Recently I’ve read an interesting webinar and a paper that using image cytometry for cell phenotyping. The idea is that, the cell segmentation algorithm can generate an output file that is analog to the FCS file format: each row represents a single-cell and columns represents the mean intensity of biomarkers (channels).

I realized that when I performed the multiplexed analysis using QuPath, the build-in algorithm can also give me the estimated intensity of each channel for each cell, which makes them great candidates for image cytometry analysis. However, I am stuck here as I am not sure how to exploit the QuPath output for FCS Express software. Obviously, FCS Express requires a specific file format - FCS to perform the image cytometry. I noticed that in Pete’s blog, there is a ShinnyApp that seems do the job, link:https://gabrielascui.shinyapps.io/qupath_to_fcs/
But when I upload the measurement file and eliminate the columns with names as requested, the system told me there is an error and I cannot download the FCS file and I am not able to see the dot plot.
I am expecting something like this eventually:


(source: PMID: 28380359)
Anyone has some exprience with this topic? Thank you in advance!

I don’t currently have a copy of FCS Express, but I was able to export the CSV file for detections from QuPath and load it into FCS Express sometime last year when I had a demo version. It did not require a specific file format at that time; I am not sure why that would have changed. You can also do something similar with CytoMAP, which has a post on the forum.

I never did get around to taking the output from FCS Express to load back into QuPath, which seemed like the greater issue.

It is important to note that QuPath does not have a default measurement that indicates the total signal for a given channel, so you might want to calculate that per cell before using software like this. Mean signals aren’t terribly representative of expression levels in many cases.


The same thing applies to whole cell measurements. And reducing the cell expansion to reduce cell overlap comes with its own problems in this context!

I realize this is a bit late, but you may also be interested in HistoCAT or CtyoMAP. I have a writeup here on transferring data back and forth between CytoMAP and QuPath.