load_cellranger_data: Make monocle3 cell_data_set using cellranger h5 Depracated -...

In scfurl/viewmaster: This package aims annotate scRNAseq data according to a reference

Make monocle3 cell_data_set using cellranger h5 Depracated - see 'load_cellranger_data' and use data_type = 'h5' Make monocle3 cell_data_set using cellranger

Description

This function reads a vector of cellranger "out" folders for conversion to a monocle3 cell_data_set object

#' There are two modes of this function that enable reading data: 1. folder-mode - Loading cellranger output folder(s) specifying folder locations using the 'folders' argument 2. file-mode - Specifying h5 files directly using 'files' argument

There are two datatypes that can be read from cellranger output: 1. Loading the cellranger h5 files either by specifying cellranger output folder(s) (containing the h5 files as geneerated by cellranger) or or can directly import .h5 files using the 'file' argument. 2. Loading the traditional 3 files output from cellranger: (matrix.mtx, barcodes.tsv, genes.tsv)

As a default, this function will use the cellranger-imposed filters for identification of droplets that do not contain cells, however, optionally, this function can return the unfiltered data as well for further manipulation (ie manual adjustment of droplet thresholds) if desired.

Usage

load_cellranger_data(
  folders = NULL,
  files = NULL,
  data_type = c("h5", "files3"),
  samplenames = NULL,
  unfiltered = F,
  unfilt_files = NULL,
  empty.droplet.threshold = 15,
  expressed_genes = TRUE,
  cell_min = 1,
  aggregated = F,
  chemistry = "SC3Pv3",
  atac_feature = "peaks"
)

Arguments

folders	A vector of cellranger folders (full folder name). These must contain an "outs" subfolders with two files; 1) raw_feature_bc_matrix.h5, and 2) filtered_feature_bc_matrix.h5 (if using the data_type = 'h5' setting, or the 3 files: matrix.mtx, barcodes.tsv, genes.tsv. For aggregated samples nrun with cellrangers 'aggregate' function, this folder should also contain the aggregation.csv file.
files	A vector of filtered h5 cellranger files. Not compatible with aggregated mode.
data_type	Use 'h5' for h5 import or 'files3' for the traditional 3 files (matrix.mtx, barcodes.tsv, genes.tsv).
samplenames	An optional vector that corresponds to the names you would like to give to each element in filelist. This defaults to the the basename of the filelist element i.e.
unfiltered	This parameter, if TRUE, returns a list containing 1) the filtered cds, 2) a list of unfiltered cds for each sample in 'folders', ultimately enabling the user to set desired cutoffs for inclusion of droplets as 'cells'.
unfilt_files	A vector of unfiltered h5 cellranger files. Not compatible with aggregated mode. Will be ignored if unfiltered is FALSE.
empty.droplet.threshold	minimum number of umis per droplet to include in unfiltered output (default 15)
expressed_genes	If true, this option removes genes that do not have expression in at least a minumum number of cells (cell_min parameter)
cell_min	Minimum number of cells that a gene needs to be expressed for the gene to be included
aggregated	will read a cellranger aggregate folder, this is currently only supported for 1 folder. The 'aggregation.csv' file must be included in the 'outs' folder.

Value

a cell_data_set object or a list of items if unfiltered data is returned (see unfiltered)

scfurl/viewmaster documentation built on Nov. 22, 2022, 2:59 p.m.