Read in 10x h5 ADT data

Helper function to load in ADT counts from h5. Leverages Rust under the hood for faster filtering and reading.

Usage

read_tenx_h5_adt(f_path, feature_type = "Antibody Capture")

Arguments

f_path

String. File to the h5 file from which to read the ADT counts.

feature_type

String. The feature type to return. Defaults here to "Antibody Capture"

Value

A dense matrix of cells x features

Examples

# pull the antibody capture layer out of a multi-modal 10x file
data <- generate_single_cell_test_data(
  syn_data_params = params_sc_synthetic_data(n_cells = 200L, n_genes = 40L)
)
adt <- generate_single_cell_test_data_adt(
  params_sc_synthetic_data_adt(n_cells = 200L)
)
f_path <- tempfile(fileext = ".h5")
write_tenx_h5_sc(
  f_path = f_path,
  counts = cbind(data$counts, as(adt$counts, "RsparseMatrix")),
  barcodes = data$obs$cell_id,
  features = data.table::data.table(
    id = c(data$var$gene_id, colnames(adt$counts)),
    name = c(data$var$ensembl_id, colnames(adt$counts)),
    feature_type = rep(
      c("Gene Expression", "Antibody Capture"),
      c(ncol(data$counts), ncol(adt$counts))
    )
  )
)
adt_counts <- read_tenx_h5_adt(f_path)
dim(adt_counts)
#> [1] 200  15

unlink(f_path)