Different features in new layer data than already exists for counts

[vertesy]

Hello,

This is a 2-in-1 issue (A, B).

I am trying to rename genes in Seurat. I know you principally disagree, however I need it.

For that end, I:

obj <- SetAssayData(obj, layer = layer.name, new.data = matrix_n) where matrix_n contains e.g. a single gene renamed.

Also tried SeuratObject::LayerData(assayobj, layer = layer.name) <- matrix_n , but same error

Expected

Matrix replaced.

Observed

1. Warning

Warning: Different features in new layer data than already exists for counts

2. Really scary behavior: it omits the gene renamed! so nrow becomes n-1!
   (A) At least this should be a "nrow !=" type of error! )

B) Question

1. How can I rename genes in freshly created object in Seurat v5?
2. I realized that names are also stored in obj@assays$RNA@features@.Data. Is there any other place that needs replacement? I guess if I replace rownames of @features@.Data, SetAssayData wont complain?

[SMP7UC]

Update: I could solve the problem by splitting object including scale.data as well, which was not included by default.

I am also getting a similar error message while scaling the data.