read10XH5: Read 10X HDF5 file

In rliger: Linked Inference of Genomic Experimental Relationships

Read 10X HDF5 file

Description

Read count matrix from 10X CellRanger HDF5 file. By default, read10XH5 load scRNA, scATAC or multimodal data into memory (inMemory = TRUE). To use LIGER in delayed mode for handling large datasets, set inMemory = FALSE to load the data as a DelayedArray object. The delayed mode only supports scRNA data for now.

Usage

read10XH5(filename, inMemory = TRUE, useNames = TRUE, featureMakeUniq = TRUE)

read10XH5Mem(filename, useNames = TRUE, featureMakeUniq = TRUE)

read10XH5Delay(filename, useNames = TRUE, featureMakeUniq = TRUE)

Arguments

filename	Character string, path to the HDF5 file.
inMemory	Logical, whether to load the data into memory. Default TRUE. FALSE loads the data as a DelayedArray object.
useNames	Logical, whether to use gene names as row names. Default TRUE. FALSE uses gene IDs instead.
featureMakeUniq	Logical, whether to make gene names unique. Default TRUE.

Value

A sparse matrix when only using older CellRanger output HDF5 file or when only one genome and one modality is detected. When multiple genomes are available, will return a list for each genome. When using multimodal data, each genome will be a list of matrices for each modality. The matrix will be of dgCMatrix class when in memory, or a TENxMatrix object when in delayed mode.

Examples

matrix <- read10XH5(
    filename = system.file("extdata/ctrl.h5", package = "rliger"),
    inMemory = TRUE
)
class(matrix) # Should show dgCMatrix
if (requireNamespace("HDF5Array", quietly = TRUE)) {
   matrix <- read10XH5(
      filename = system.file("extdata/ctrl.h5", package = "rliger"),
      inMemory = FALSE
   )
   print(class(matrix)) # Should show TENxMatrix
}

rliger documentation built on June 8, 2025, 1:56 p.m.