Error: useNames = NA is defunct. Instead, specify either useNames = TRUE or useNames = FALSE. Traceback:

[ArzooShamoon]

I tried to run singleR for cell type annotation to each cluster.
head(Sample.combined)
A data.frame: 10 × 6
orig.ident nCount_RNA nFeature_RNA percent.mt integrated_snn_res.0.5 seurat_clusters

MSC_AAACGAAAGAGAATCT-1 MSC 19477 4423 3.9533809 0 0
MSC_AAACGAAAGAGTCCGA-1 MSC 8237 3180 2.3188054 10 10
MSC_AAACGAACACAGCATT-1 MSC 4225 1859 7.3609467 4 4
MSC_AAACGCTCAAAGACGC-1 MSC 12958 3852 2.5158203 12 12
MSC_AAACGCTGTAACTAAG-1 MSC 874 294 0.5720824 1 1
MSC_AAAGAACGTACACTCA-1 MSC 6487 1351 1.0790812 1 1
MSC_AAAGGATTCTGTGCAA-1 MSC 1591 822 4.2740415 0 0
MSC_AAAGGGCAGCGTCAGA-1 MSC 27910 5472 3.2891437 0 0
MSC_AAAGGGCAGGAGAGGC-1 MSC 4829 995 0.5384138 9 9
MSC_AAAGGGCAGGTACCTT-1 MSC 18817 4437 2.4711697 0 0

ref <- celldex::HumanPrimaryCellAtlasData()
colData(ref)
results <- SingleR (test = as.SingleCellExperiment(Sample.combined), ref = ref, labels = ref$label.main)
results

Error: useNames = NA is defunct. Instead, specify either useNames = TRUE or useNames = FALSE.
Traceback:

SingleR(test = as.SingleCellExperiment(Sample.combined), ref = ref,
. labels = ref$label.main)
trainSingleR(ref, labels, genes = genes, sd.thresh = sd.thresh,
. de.method = de.method, de.n = de.n, de.args = de.args, aggr.ref = aggr.ref,
. aggr.args = aggr.args, recompute = recompute, restrict = restrict,
. check.missing = FALSE, BNPARAM = BNPARAM, num.threads = num.threads,
. BPPARAM = BPPARAM)
mapply(FUN = .identify_genes, ref = ref, labels = labels, genes = genes,
. MoreArgs = list(de.method = de.method, de.n = de.n, de.args = de.args,
. BPPARAM = BPPARAM), SIMPLIFY = FALSE)
(function (ref, labels, genes = "de", de.method = "classic",
. de.n = NULL, de.args = list(), BPPARAM = BPPARAM)
. {
. if (length(labels) != ncol(ref)) {
. stop("number of labels must be equal to number of cells")
. }
. if (.is_list(genes)) {
. is.char <- vapply(genes, is.character, TRUE)
. if (all(is.char)) {
. genes <- .convert_per_label_set(genes)
. }
. else if (any(is.char)) {
...
matrixStats::rowMedians(x, rows = rows, cols = cols, na.rm = na.rm,
. dim. = dim., ..., useNames = useNames)
deprecatedUseNamesNA()
.Defunct(msg = "useNames = NA is defunct. Instead, specify either useNames = TRUE or useNames = FALSE.",
. package = .packageName)

[LTLA]

Make sure your packages are up to date, in particular MatrixGenerics, sparseMatrixStats and DelayedMatrixStats; I can't remember exactly which one of them was passing along useNames=NA, but it should have been fixed now.

[adRn-s]

HenrikBengtsson/matrixStats#246

This comes from matrixStats, downgrading should fix it temporarily:

remotes::install_version("matrixStats", version="1.1.0") # restart your session and run previous scripts

The default value of their functions now is useNames=TRUE. I am yet to find a changelog... this is so annoying.

[CakeManka]

"Make sure your packages are up to date, in particular MatrixGenerics, sparseMatrixStats and DelayedMatrixStats; I can't remember exactly which one of them was passing along useNames=NA, but it should have been fixed now."
this is a fasle means to handle error.
changing the raw code is a correct manner.

[CakeManka]

Make sure your packages are up to date, in particular MatrixGenerics, sparseMatrixStats and DelayedMatrixStats; I can't remember exactly which one of them was passing along useNames=NA, but it should have been fixed now.

i think there are a question in the raw code. in the fact, i can not use this package just only change the version of MatrixGenerics, sparseMatrixStats and DelayedMatrixStats.

[adRn-s]

HenrikBengtsson/matrixStats#246

@CakeManka Please check that link, I got answers from core maintainers. They're aiming to fix it in the new release. Apparently, the situation is extremely complex and takes time.