I would like to annotate or label UMAP clusters with cell types in a multiple myeloma single-cell RNA-seq dataset using the scRNAseq R package in conjunction with the SingleR package, if possible. I found a dataset that I would like to use on the scRNAseq dataset collection which is in the LedergorMyelomaData() function which returns a SingleCellExperiment object. I notice that there is no ''cell label'' column within this object? Does that mean I cannot use this data as a reference to annotate my own myeloma dataset's clusters? In terms of cell labels, does it depend on each dataset in the scRNAseq collection if it has the labels there or not? I also searched through the ExperimentHub() repository and found the same dataset which seems to be under the code EH6694 and is shown in the below screenshot.

If, it is not possible to use the Ledergor data for cell type annotation, what dataset would you recommend using instead? I would like to compare to a public reference if possible.

library(scRNAseq)
#
mm_annotations <- LedergorMyelomaData()
#
head(mm_annotations@colData)

In terms of cell labels, does it depend on each dataset in the scRNAseq collection if it has the labels there or not?

Yes, it basically depends on whether the authors of the original study have provided the cell type labels. You can exactly see how the resource was created by reading the relevant script, e.g., on GitHub via https://github.com/LTLA/scRNAseq/blob/master/inst/scripts/2.8.0/make-ledergor-myeloma-data.Rmd or from within R by opening the file given by the value of this R code system.file("scripts/2.8.0/make-ledergor-myeloma-data.Rmd", package = "scRNAseq").

You could double-check if the authors of the original study provided the cell labels somewhere in the original publication, but I'm guessing they haven't.

If, it is not possible to use the Ledergor data for cell type annotation, what dataset would you recommend using instead?

There may be people here who can provide suggestions, but it's really up to you and the people you are working with to decide this because it's about identifying a biologically relevant reference dataset for your data (and none of us know your data as well as you do!).