data_process: Data process

In keyalone/scDEA: Performing differential expression analysis in single cell RNA sequencing data via ensemble learning

Description

This function focus on dealing various single-cell RNA-seq input and unifying output format.

Usage

data_process(Data, group, norm.form = "CPM", is.normalized = FALSE)

Arguments

Data	single-cell RNA-seq matrix. The format could be raw-counts, FPKM/RPKM, TPM or UMI-counts. The matrix need include gene names and cell names.
group	group information. The cell need be divided into two category.
norm.form	character item. We provide several normalized method for raw-counts data. The method include "TMM","RLE", "CPM". The default is "CPM".
is.normalized	logical. A logical flag to determin whether or not the input dataset normalizes. If TRUE, we will take the Data as normcounts and input for downstream analysis. If not, we provide method for the process.

Details

We take relative2abs transfering relative expression values into absolute transcript counts. However, the process maybe break the original dataset statistical properties. Hence, we advise user don't normalize firstly.

Value

• sce : A SingleCellExperiment item. The object include expression matrix, group information. The expression matrix contains counts and normcounts.

Author(s)

Huisheng, Li, <[e-mail]>

Examples

data("Grun.counts.hvg")
data("Grun.group.information")
sce <- data_process(Data = Grun.counts.hvg, group = Grun.group.information)

keyalone/scDEA documentation built on Dec. 21, 2021, 6:36 a.m.