bootstrap_enrichment_test: Bootstrap cell type enrichment test

In NathanSkene/EWCE: Expression Weighted Celltype Enrichment

Bootstrap cell type enrichment test

Description

bootstrap_enrichment_test takes a genelist and a single cell type transcriptome dataset and determines the probability of enrichment and fold changes for each cell type.

Usage

bootstrap_enrichment_test(
  sct_data = NULL,
  hits = NULL,
  bg = NULL,
  genelistSpecies = NULL,
  sctSpecies = NULL,
  sctSpecies_origin = sctSpecies,
  output_species = "human",
  method = "homologene",
  reps = 100,
  no_cores = 1,
  annotLevel = 1,
  geneSizeControl = FALSE,
  controlledCT = NULL,
  mtc_method = "BH",
  sort_results = TRUE,
  standardise_sct_data = TRUE,
  standardise_hits = FALSE,
  verbose = TRUE,
  localHub = FALSE,
  store_gene_data = TRUE
)

Arguments

sct_data	List generated using generate_celltype_data.
hits	List of gene symbols containing the target gene list. Will automatically be converted to human gene symbols if geneSizeControl=TRUE.
bg	List of gene symbols containing the background gene list (including hit genes). If bg=NULL, an appropriate gene background will be created automatically.
genelistSpecies	Species that hits genes came from (no longer limited to just "mouse" and "human"). See list_species for all available species.
sctSpecies	Species that sct_data is currently formatted as (no longer limited to just "mouse" and "human"). See list_species for all available species.
sctSpecies_origin	Species that the sct_data originally came from, regardless of its current gene format (e.g. it was previously converted from mouse to human gene orthologs). This is used for computing an appropriate backgrund.
output_species	Species to convert sct_data and hits to (Default: "human"). See list_species for all available species.
method	R package to use for gene mapping: "gprofiler" : Slower but more species and genes. "homologene" : Faster but fewer species and genes. "babelgene" : Faster but fewer species and genes. Also gives consensus scores for each gene mapping based on a several different data sources.
reps	Number of random gene lists to generate (Default: 100, but should be >=10,000 for publication-quality results).
no_cores	Number of cores to parallelise bootstrapping reps over.
annotLevel	An integer indicating which level of sct_data to analyse (Default: 1).
geneSizeControl	Whether you want to control for GC content and transcript length. Recommended if the gene list originates from genetic studies (Default: FALSE). If set to TRUE, then hits must be from humans.
controlledCT	[Optional] If not NULL, and instead is the name of a cell type, then the bootstrapping controls for expression within that cell type.
mtc_method	Multiple-testing correction method (passed to p.adjust).
sort_results	Sort enrichment results from smallest to largest p-values.
standardise_sct_data	Should sct_data be standardised? if TRUE: When sctSpecies!=output_species the sct_data will be checked for object formatting and the genes will be converted to the orthologs of the output_species with standardise_ctd (which calls map_genes internally). When sctSpecies==output_species, the sct_data will be checked for object formatting with standardise_ctd, but the gene names will remain untouched.
standardise_hits	Should hits be standardised? If TRUE: When genelistSpecies!=output_species, the genes will be converted to the orthologs of the output_species with convert_orthologs. When genelistSpecies==output_species, the genes will be standardised with map_genes. If FALSE, hits will be passed on to subsequent steps as-is.
verbose	Print messages.
localHub	If working offline, add argument localHub=TRUE to work with a local, non-updated hub; It will only have resources available that have previously been downloaded. If offline, Please also see BiocManager vignette section on offline use to ensure proper functionality.
store_gene_data	Store sampled gene data for every bootstrap iteration. When the number of bootstrap reps is very high (>=100k) and/or the number of genes in hits is very high, you may want to set store_gene_data=FALSE to avoid using excessive amounts of CPU memory.

Value

A list containing three elements:

• hit.cells: vector containing the summed proportion of expression in each cell type for the target list.

• gene_data: data.table showing the number of time each gene appeared in the bootstrap sample.

• bootstrap_data: matrix in which each row represents the summed proportion of expression in each cell type for one of the random lists

• controlledCT: the controlled cell type (if applicable)

Examples

# Load the single cell data
sct_data <- ewceData::ctd()
# Set the parameters for the analysis
# Use 3 bootstrap lists for speed, for publishable analysis use >=10,000
reps <- 3
# Load gene list from Alzheimer's disease GWAS
hits <- ewceData::example_genelist()

# Bootstrap significance test, no control for transcript length or GC content
full_results <- EWCE::bootstrap_enrichment_test(
    sct_data = sct_data,
    hits = hits,
    reps = reps,
    annotLevel = 1,
    sctSpecies = "mouse",
    genelistSpecies = "human")

NathanSkene/EWCE documentation built on April 30, 2024, 5:21 p.m.