LRcellCore: Find most enriched cell types in bulk DE genes by Logistic...

In marvinquiet/LRcell: Differential cell type change analysis using Logistic/linear Regression

Find most enriched cell types in bulk DE genes by Logistic Regression

Description

This is a function which takes marker genes from single-cell RNA-seq as reference to calculate the enrichment of certain cell types in bulk DEG analysis. We assume that bulk DEG is derived from certain cell-type specific pattern.

Usage

LRcellCore(gene.p, marker.g, method, min.size = 5, sig.cutoff = 0.05)

Arguments

gene.p	Named vector of gene-level pvalues from DEG analysis, i.e. DESeq2, LIMMA
marker.g	List of Cell-type specific marker genes derived from single-cell RNA-seq. The name of the list is cell-type or cluster name, the values are marker genes vectors or numeric named vectors. LRcell provides marker genes list in different human/mouse brains, but users could use their own marker gene list as input. default: NULL
method	Either 'logistic regression' or 'linear regression'. Logistic regression equally treats cell-type specific marker genes, however, if certain values could determine the importance of marker genes, linear regression can be performed, default: LR.
min.size	Minimal size of a marker gene set, will impact the balance of labels
sig.cutoff	Cutoff for input genes' pvalues, default: 0.05.

Value

A dataframe of LRcell statistics as described in LRcell.

Examples

data(mouse_FC_marker_genes)
data(example_gene_pvals)
res <- LRcellCore(example_gene_pvals, mouse_FC_marker_genes, method="LR")

marvinquiet/LRcell documentation built on Sept. 16, 2022, 9:09 a.m.