I'm trying to run classifySingleR function from SingleR package using MouseRNAseqData() as a reference to classify the cells from a Seurat dataset that has not yet been published, but I got the following error:

Error in h(simpleError(msg, call)) : error in evaluating the argument 'x' in selecting a method for function 'colRanks': error in evaluating the argument 'seed' in selecting a method for function 'DelayedArray': invalid or not-yet-implemented 'Matrix' subsetting

library(SingleR)

df_genes <- readRDS("Seuratfile.rds")
mrnaseq <- MouseRNAseqData()
test <- df_genes@assays$RNA@data # a Seurat dataset
shared_markers <- row.names(mrnaseq)[row.names(mrnaseq)%in%row.names(test)] # shared markers in the reference and in our gene expression data
sub_test <- test[shared_markers,]

singler_out <- classifySingleR(test = sub_test, trained = mrnaseq) 

sessionInfo( )