RunCellHGT: Run HyperGeometric Test on cells

In RausellLab/CelliD: Unbiased Extraction of Single Cell gene signatures using Multiple Correspondence Analysis

Run HyperGeometric Test on cells

Description

RunCellHGT calculates the gene signatures for each cells and performs hypergeometric test against a user defined gene signatures/pathways (named list of genes). It returns a score of enrichment in the form of -log10 pvalue(see log.trans argument). The obtained matrix can then be integrated in Seurat or SingleCellExperiment object. It can notably be used with cell type signatures to predict cell types or with functionnal pathways

Usage

RunCellHGT(
  X,
  pathways,
  reduction,
  n.features,
  features,
  dims,
  minSize,
  log.trans,
  p.adjust
)

## S3 method for class 'SingleCellExperiment'
RunCellHGT(
  X,
  pathways,
  reduction = "MCA",
  n.features = 200,
  features = NULL,
  dims = seq(50),
  minSize = 10,
  log.trans = TRUE,
  p.adjust = TRUE
)

## S3 method for class 'Seurat'
RunCellHGT(
  X,
  pathways,
  reduction = "mca",
  n.features = 200,
  features = NULL,
  dims = seq(50),
  minSize = 10,
  log.trans = TRUE,
  p.adjust = TRUE
)

Arguments

X	Seurat or SingleCellExperiment object with mca performed
pathways	geneset to perform hypergeometric test on (named list of genes)
reduction	name of the MCA reduction
n.features	integer of top n features to consider for hypergeometric test
features	vector of features to calculate the gene ranking by default will take everything in the selected mca reduction.
dims	MCA dimensions to use to compute n.features top genes.
minSize	minimum number of overlapping genes in geneset and
log.trans	if TRUE tranform the pvalue matrix with -log10 and convert it to sparse matrix
p.adjust	if TRUE apply Benjamini Hochberg correction to p-value

Value

a matrix of benjamini hochberg adjusted pvalue pvalue or a sparse matrix of (-log10) benjamini hochberg adjusted pvalue

Examples

seuratPbmc <- RunMCA(seuratPbmc, nmcs = 5)
Enrichment <- RunCellHGT(X = seuratPbmc, pathways = Hallmark, dims = 1:5)

RausellLab/CelliD documentation built on Jan. 12, 2024, 3:44 a.m.