gseaScores: Compute enrichment scores for GSEA (Gene Set Enrichment...

In HTSanalyzeR: Gene set over-representation, enrichment and network analyses for high-throughput screens

Description

'gseaScores' computes the enrichment score, running sum scores and positions of hits for GSEA on one gene set.

'gseaScoresBatch' computes enrichment scores for both input 'geneList' and its permutations for GSEA on one gene set.

'gseaScoresBatchParallel' computes enrichment scores for both input 'geneList' and their permutations for GSEA on multiple gene sets in parallel.

Usage

gseaScores(geneList, geneSet, exponent=1, mode = "score")

gseaScoresBatch(geneList, geneNames.perm, geneSet, exponent=1,
nPermutations=1000)

gseaScoresBatchParallel(geneList, geneNames.perm, collectionOfGeneSets,
exponent=1, nPermutations=1000)

Arguments

geneList	a numeric or integer vector of phenotypes in descending or ascending order with elements named by their gene identifiers (no duplicates nor NA values)
geneNames.perm	a character matrix including permuted gene identifiers of input 'geneList'. The first column of this matrix should be the gene identifiers of 'geneList', while other columns are vectors of permuted gene identifiers.
geneSet	a character vector specifying a gene set (no names, just a vector of characters corresponding to the IDs)
collectionOfGeneSets	a list of gene sets. Each gene set in the list is a character vector of gene identifiers
exponent	a single integer or numeric value used in weighting phenotypes in GSEA
nPermutations	a single integer or numeric value specifying the number of permutations for deriving p-values in GSEA
mode	a single character value specifying to return only a score (if set as "score"), or all the necessary elements to make a plot (if "graph") (see "gseaPlot" for more details).

Details

The type of identifiers used in the gene sets and gene list must obviously match.

Value

* gseaScores will return:

enrichmentScore	a single numeric value of the enrichment score
runningScore	a numeric vector of running sum scores (only in mode "graph")
positions	a numeric vector of positions in the ranked phenotype vector of the genes in the gene set (only in mode "graph")

* gseaScoresBatch will return a list consisting of:

scoresObserved	a single numeric value of the enrichment score for input 'geneList'
scoresperm	a numeric vector of enrichment scores for permutation tests

* gseaScoresBatchParallel will return a matrix, in which one column is the 'gseaScoresBatch' result for each gene set.

Author(s)

Xin Wang, Camille Terfve

References

Subramanian, A., Tamayo, P., Mootha, V. K., Mukherjee, S., Ebert, B. L., Gillette, M. A., Paulovich, A., Pomeroy, S. L., Golub, T. R., Lander, E. S. & Mesirov, J. P. (2005) Gene set enrichment analysis: A knowledge-based approach for interpreting genome-wide expression profiles. Proc. Natl. Acad. Sci. USA 102, 15545-15550.

Examples

##example 1
gl <- runif(100, min=0, max=5)
gl <- gl[order(gl, decreasing=TRUE)]
names(gl) <- as.character(sample(x=seq(from=1, to=100, by=1), size=100,
replace=FALSE))
gs <- sample(names(gl), size=20, replace=FALSE)
##GSEA for input phenotype vector on a single gene set
gsea <- gseaScores(geneList=gl, geneSet=gs, mode="score", exponent=1)
##GSEA for both input phenotype vector and permutation tests on a single gene set
nPermutations <- 100
glPerm <- sapply(1:nPermutations, function(n) names(gl)[sample(1:length(gl),
length(gl), replace=FALSE)])
glPerm <- cbind(names(gl), glPerm)
gseaBatch<-gseaScoresBatch(geneList=gl, geneNames.perm=glPerm, geneSet=gs,
nPermutations=100, exponent=1)
##example 2
## Not run: 
library(KEGG.db)
library(org.Dm.eg.db)
library(snow)
##load phenotype vector (see the vignette for details about the
##preprocessing of this data set)
data("KcViab_Data4Enrich")
DM_KEGG <- KeggGeneSets(species="Dm")
##GSEA for input gene list on a single gene set
test <- gseaScores(geneList=KcViab_Data4Enrich, geneSet=DM_KEGG[[1]], 
exponent=1, mode="graph")
##GSEA for both input gene list and permutation tests on multiple gene
##sets in parallel
nPermutations <- 100
glPerm <- sapply(1:nPermutations, function(n) names(KcViab_Data4Enrich)[
sample(1:length(KcViab_Data4Enrich), length(KcViab_Data4Enrich), 
replace=FALSE)])
glPerm<-cbind(names(KcViab_Data4Enrich), glPerm)
options(cluster=makeCluster(4,"SOCK"))
gseaBatchPar <- gseaScoresBatchParallel(geneList=KcViab_Data4Enrich,
geneNames.perm=glPerm, collectionOfGeneSets=DM_KEGG[1:10], nPermutations=100,
exponent=1)
if(is(getOption("cluster"),"cluster")) {
	stopCluster(getOption("cluster"))
	options(cluster=NULL)
}

## End(Not run)

HTSanalyzeR documentation built on Oct. 31, 2019, 7:10 a.m.