Storing a custom dimensional reduction Seurat v5

[mrmxhmd]

Hello,

I was following this vignette to add a custom dimensional reduction with Seurat v5, however I am encountering an error.

The table I use is: Combined_embedding_v0.csv

And here is my code

pia <- read.csv("Combined_embedding_v0.csv")
names(pia) <- NULL
colnames(pia)[2:21] <- paste0("pia_", 1:20)
pia <- subset (pia, select = 2:21,)
pia <- data.matrix(pia, rownames.force = NA)
fsMerge[["pia"]] <- CreateDimReducObject(embeddings = pia, key = "pia_", assay = DefaultAssay(fsMerge))

This is the error I receive: Error: Cannot add new cells with [[<-

I also tried going through the test data in the vignette and I get this error at the step d <- dist(t(GetAssayData(pbmc, slot = "scale.data"))):
Error in h(simpleError(msg, call)) :
error in evaluating the argument 'x' in selecting a method for function 't': no applicable method for 'GetAssayData' applied to an object of class "seurat"

Is this a problem with the version or am I doing something wrong?

Thank you in advance!

[jsicherman]

Unsure if this is the now-intended way or not, but you could effectively "override" the SeuratObject's implementation of [[ by directly indexing into the object's reductions slot, like so (note the new @reductions):

fsMerge@reductions[["pia"]] <- CreateDimReducObject(embeddings = pia, key = "pia_", assay = DefaultAssay(fsMerge))