var_filter_counts: Filter low-variance genes
In mjdufort/countSubsetNorm: RNAseq counts subsetting and normalizing.

This function filters a counts object, keeping the n_keep genes with the highest variance.

1
2
3

var_filter_counts(
  counts, counts, n_keep=8000, genes="rows",
  log2_transform=FALSE)

`counts`	a matrix or data frame of gene expression counts.
`n_keep`	integer, the maximum number of genes to retain
`genes`	character string, the dimension of `counts` containing the genes. Accepted values are "rows", "columns", and partial matches. Defaults to "rows"
`log2_transform`	boolean, whether to log2 transform the counts prior to variance filtering. If the counts have not already been transformed, should be set to `TRUE`. Defaults to `FALSE`.