testUnidirectionality: Test gene sets for unidirectional enrichment

In saeyslab/triwise: Gene expression changes in three biological conditions

Description

Test gene sets for unidirectional enrichment

Usage

testUnidirectionality(barycoords, gsets, Gdiffexp = NULL,
  statistic = "diffexp", bm = NULL, minknown = 5, mindiffexp = 0,
  maxknown = 1500, mc.cores = getOption("mc.cores", default = 1), ...)

Arguments

barycoords	Dataframe containing for every gene its barycentric coordinates, as returned by transformBarycentric
gsets	List of character vectors, each containing a set of genes (gene identifiers)
Gdiffexp	Differentially expressed genes
statistic	A string denoting the measure used for the strength of upregulation of a particular gene. diffexp: Whether a gene is differentially expressed (1 versus 0) rank: The rank of the maximal log fold-change r: The maximal log fold-change z: Custom using the z-column within barycoords angle: Uses a rayleigh z-test ignoring non-differentially expressed genes within the gene set
bm	Previously calculated background model using the generateBackgroundModel
minknown	Minimal number of genes within a gene set for it to be considered for enrichment
mindiffexp	Minimal number of genes differentially expressed within a gene set for it to be considered for enrichment
maxknown	Maximal number of genes within a gene set for it to be considered for enrichment
mc.cores	Number of processor cores to use. Due to limitations of the parallel package, this does not work on Windows
...	Parameters for generateBackgroundModel

Value

Dataframe containing for every gene set which passed the filtering:

• p-value of unidirectionality

• q-value of unidirectionality (p-value corrected for multiple testing)

• average angle

Examples

Eoi = matrix(rnorm(1000*3, sd=0.5), 1000, 3, dimnames=list(1:1000, c(1,2,3)))
Eoi[1:100,1] = Eoi[1:100,1] + 4 # the first 100 genes are more upregulated in the first condition
barycoords = transformBarycentric(Eoi)

gsets = list(a=1:50, b=80:150, c=200:500)
testUnidirectionality(barycoords, gsets, Gdiffexp=(1:1000)[barycoords$r > 1])

saeyslab/triwise documentation built on May 29, 2019, 12:56 p.m.