Rename cells in Seurat

Question

[V]

Hello,

I've seen the relative thread of renaming cells, if for example merging Seurats from different 10x experiments which will mean some barcodes will be overlapping. The issue I'm having is that I want to save as Matrix the counts of 6 clusters. I am trying this with the code below :

rawdata <- as.matrix(GetAssayData(Stromal, slot = "counts")[, WhichCells(Stromal, idents = c("1", "2", "3", "4", "5", "6"), slot = "counts")])

this gives me the following error :

Error in intI(j, n = x@Dim[2], dn[[2]], give.dn = FALSE) : 
  no 'dimnames[[.]]': cannot use character indexing

which I assume means I have duplicate cell names. So I would need to somehow rename the cells based on their cluster? is there a way to do is?

Or alternatively if I've got it wrong, is there a way around the above error? Thanks

[igor]

If you have a Seurat object, you do not have duplicate cell names.

[V]

Thank you. Would you happen to know what the error means?

Answer 1

[jared.andrews07]

I think you're complicating things for yourself a bit.

# subset first
Stromal.subset <- subset(x = Stromal, idents = c("1", "2", "3", "4", "5", "6"))

# Then get counts as a matrix.
my.matrix <- as.matrix(GetAssayData(object = Stromal.subset, slot = "counts"))

[V]

Thanks for this! First part works fine and I get a Seurat object containing the clusters I want.

When I run the second command it goes through apparently alright, but I get a file that contains nothing for some reason. The file just says " num [0 , 0 ] "

Answer 2

[V]

Ok solved in a round about way. Not sure what the initial error is referring to. If anybody has any ideas what it was please let me know as the above I posted in the initial question use to work.

This is how I went about it for reference in case someone else runs into the same issue :

cells.use <- WhichCells(object = Stromal , ident = c("1", "2", "3", "4", "5", "6"))

expr <- GetAssayData(object = Stromal, assay = "RNA", slot = "counts")[, cells.use]
expr <- as(Class = 'matrix', object = expr)
write.csv(x = expr, file = "expression_cluster.csv", quote = FALSE)