search

lmweber / diffcyt

R package for differential discovery analyses in high-dimensional cytometry data
MIT License
19 stars 12 forks source link

How to use diffcyt to identify differentially-expressed proteins/biomarkers of a cluster in flow cytometry data? #50

Closed denvercal1234GitHub closed 1 year ago

denvercal1234GitHub commented 1 year ago

Hi there,

The tutorial (https://www.bioconductor.org/packages/release/bioc/vignettes/diffcyt/inst/doc/diffcyt_workflow.html) is very useful, but it only shows how to perform the analysis between conditions.

I had performed clustering using CATALYST.

Would you mind advising me how to adjust the codes, especially diffcyt::createDesignMatrix and contrast to identify differentially-expressed proteins (biomarkers) of a cluster compared to every other cluster, done for each cluster as in DEGs for scRNAseq analyses? Or is it not possible?

Thank you for your help!

markrobinsonuzh commented 1 year ago

hi @denvercal1234GitHub,

Yeah, I mean, diffcyt is methodology for performing analysis between conditions, not between clusters .. so, for finding marker genes, I suggest that you use something else. For example, you might be able to use the same findMarkers function in scran that is used in scRNA-seq marker DE detection, since the data used in CATALYST is a SingleCellExperiment object. I have actually never done this, because this is not typically the way that people do it for cytometry (generally, people pick markers to use in the first place). But, from a stats perspective, it should still work.

Cheers, Mark