Help with extracting residual variance or vst.variance.expected/vst.variance.standartised

[nFursova]

Hi,

I am very new to Seurat and I am trying to use it to calculate variance of gene expression from scRNA 10x data, in particular, residual variance or what used to be standartised vst.variance in the previous version. However, I am struggling to get the meta.data assay in seurat_object.

This is that I am trying to do:

Load scRNA-seq data frame

scRNA = read.table('file.csv', header = T, stringsAsFactors = F, sep = ',')

rownames(scRNA) = scRNA$REV

scRNA = select(scRNA, -REV)

scRNA = as.matrix(scRNA)

Perform variance stabilization using seurat -----------------------

seurat_object <- CreateSeuratObject(counts = scRNA, min.cells = 10, min.features = 200)

seurat_object <- SCTransform(seurat_object, verbose = T)

And I get seurat_object that looks like this:

RNA@meta.data and SC@meta.features are empty.

I also tried running the old method:

seurat_obj_vst <- NormalizeData(seurat_object,
normalization.method = "LogNormalize",
scale.factor = 10000)

seurat_obj_vst <- FindVariableFeatures(
seurat_obj_vst,
selection.method = "vst",
nfeatures = 2000
)
But this returns no meta.data either.

I really can't figure out if I am doing something wrong or where to extract the residual variance info from in the new Seurat version.

I am using Seurat 5.2.1.

I would really appreciate your help with issue!

[nFursova]

I figured out how to do it with v3 Seurat object:

options(Seurat.object.assay.version = "v3")

seurat_object <- CreateSeuratObject(counts = scRNA, min.cells = 10, min.features = 200)

seurat_object <- NormalizeData(seurat_object)

seurat_object <- FindVariableFeatures(seurat_object)

seurat_object <- ScaleData(seurat_object)

variance_data <- seurat_object@assays[["RNA"]]@meta.data