load_se_from_tables: load_se_from_tables

In MonashBioinformaticsPlatform/celaref: Single-cell RNAseq cell cluster labelling by reference

Description

Create a SummarizedExperiment object (dataset_se) from a count matrix, cell information and optionally gene information.

load_se_from_files is a wrapper for load_se_from_tables that will read in tables from specified files.

Usage

load_se_from_tables(counts_matrix, cell_info_table, gene_info_table = NA,
  group_col_name = "group", cell_col_name = NA)

load_se_from_files(counts_file, cell_info_file, gene_info_file = NA,
  group_col_name = "group", cell_col_name = NA)

Arguments

counts_matrix	A tab-separated matrix of read counts for each gene (row) and each cell (column). Columns and rows should be named.
cell_info_table	Table of cell information. If there is a column labelled cell_sample, that will be used as the unique cell identifiers. If not, the first column is assumed to be cell identifiers, and will be copied to a new feild labelled cell_sample. Similarly - the clusters of these cells should be listed in one column - which can be called 'group' (case-sensitive) or specified with group_col_name. Minimal data format: <cell_sample> <group>
gene_info_table	Optional table of gene information. If there is a column labelled ID, that will be used as the gene identifiers (they must be unique!). If not, the first column is assumed to be a gene identifier, and will be copied to a new feild labelled ID. Must match all rownames in counts_matrix. If omitted, ID wll be generated from the rownames of counts_matrix. Default=NA
group_col_name	Name of the column in cell_info_table containing the cluster/group that each cell belongs to. Case-sensitive. Default='group'
cell_col_name	Name of the column in cell_info_table containing a cell id. Ignored if cell_sample column is already present. If omitted, (and no cell_sample column) will use first column. Case-sensitive. Default=NA
counts_file	A tab-separated file of a matrix of read counts. As per counts_matrix. First column should be gene ID, and top row cell ids.
cell_info_file	Tab-separated text file of cell information, as per cell_info_table. Columns must have names.
gene_info_file	Optional tab-separated text file of gene information, as per gene_info_file. Columns must have names. Default=NA

Details

This function makes a SummarizedExperiment object in a form that should work for celaref functions. Specifically, that means it will have an 'ID' feild for genes (view with rowData(dataset_se)), and both 'cell_sample' and 'group' feild for cells (view with colData(dataset_se)). See parameters for detail. Additionally, the counts will be an integer matrix (not a sparse matrix), and the group feild (but not cell_sample or ID) will be a factor.

Note that data will be subsetted to cells present in both the counts matrix and cell info, this is handy for loading subsets of cells. However, if gene_info_file is defined, all genes must match exactly.

The load_se_from_files form of this function will run the same checks, but will read everything from files in one go. The load_se_from_tables form is perhaps more useful when the annotations need to be modified (e.g. programmatically adding a different gene identifier, renaming groups, removing unwanted samples).

Note that the SummarizedExperiment object can also be created without using these functions, it just needs the cell_sample, ID and group feilds as described above. Since sometimes it might be easier to add these to an existing SummarizedExperiment from upstream analyses.

Value

A SummarisedExperiment object containing the count data, cell info and gene info.

Functions

• load_se_from_files: To read from files

See Also

SummarizedExperiment For general doco on the SummarizedExperiment objects.

Other Data loading functions: contrast_each_group_to_the_rest_for_norm_ma_with_limma, load_dataset_10Xdata

Examples

# From data frames (or a matrix for counts) :
demo_se <- load_se_from_tables(counts_matrix=demo_counts_matrix, 
                               cell_info_table=demo_cell_info_table)
demo_se <- load_se_from_tables(counts_matrix=demo_counts_matrix, 
                               cell_info_table=demo_cell_info_table, 
                               gene_info_table=demo_gene_info_table)

# Or from data files : 
counts_filepath    <- system.file("extdata", "sim_query_counts.tab",    package = "celaref")
cell_info_filepath <- system.file("extdata", "sim_query_cell_info.tab", package = "celaref")
gene_info_filepath <- system.file("extdata", "sim_query_gene_info.tab", package = "celaref")

demo_se <- load_se_from_files(counts_file=counts_filepath, cell_info_file=cell_info_filepath)
demo_se <- load_se_from_files(counts_file=counts_filepath, cell_info_file=cell_info_filepath, 
                              gene_info_file=gene_info_filepath )

MonashBioinformaticsPlatform/celaref documentation built on June 5, 2019, 11:35 a.m.