Running FindAllMarkers after SCtransform on Xenium data

[billiep1234]

Hi,

I am looking to analyse Xenium data through cluster analysis using SCtransform. The workflow I am using is based on this vignette (https://satijalab.org/seurat/articles/seurat5_spatial_vignette_2#mouse-brain-10x-genomics-xenium-in-situ) for Xenium, however I am skipping some parts of the vignette as I don't have the reference data to do so (mainly in creating the imagedimplots). I can successfully create a dimplot after umap, however when I run findallmarkers I encounter some problems.

My workflow after reading and subsetting a merged RDS file is:
object <- SCTransform(object, assay = "Xenium"
object <- RunPCA(object, npcs = 30, features = rownames(object))
object <- FindClusters(object, resolution = 0.3)
object <- RunUMAP(object, dims = 1:30)
object_markers <- FindAllMarkers(object, assay = "SCT", only.pos = TRUE, min.pct = 0.25, logfcthreshold = 0.25, text.use = "negbinom"

When FindAllMarkers is run I get the warning: When testing 0 versus all: object contains multiple models with unequal library sizes. Run 'PrepSCTFindMarkers()' before running 'FindMarkers()', for each of the clusers
I also get the warning: No DE genes identified

After object <- PrepSCTFindMarkers(object, assay = "SCT"), I get the error: Found 13 SCT models. Recorrecting SCT counts using minimum median counts: 85
Error in PrepSCTFindMarkers(object, assay "SCT"): Multiple UMI assays are used for the SCtransform: Xenium, RNA

If anyone has had this issue before, I would appreciate any advice! Thank you