GSReg_GeneSets_VReg.: Performs Gene Set Analysis using Expression Variation...

Description Usage Arguments Value Author(s) See Also Examples

Description

GSReg.GeneSets.EVA performs modified version DIRAC papers. Using a theoretical analysis, we can calculate p-value which makes extreme low p-values available.

Usage

1
 GSReg.GeneSets.EVA(geneexpres, pathways, phenotypes,minGeneNum=5) 

Arguments

geneexpres

the matrix of gene expressions. The rownames must represent gene names and the columns represent samples. There must not be any missing values. Please use imputation or remove the genes with missing values.

pathways

a list containing pathway information. Each element represents a pathway as a character vector. The genes shown in the pathway must be present in geneexpres. geneexpres must have numeric and finite numbers.

phenotypes

a binary factor containing the phenotypes for samples in geneexpres; hence, the column number of geneexpres and the length of phenotypes must be equal.

minGeneNum

the minimum number of genes required in a pathway.

Value

a list of analysis for all pathways.

$E1

the modified variance on the pathway within the samples from levels(phenotypes)[1].

$E2

the modified variance on the pathway within the samples from levels(phenotypes)[2].

$E12

the modified variance on the pathway across the samples from levels(phenotypes)[1] tolevels(phenotypes)[2].

$VarEta1

the estimation of the modified variance on the pathway within the samples from levels(phenotypes)[1].

$VarEta2

the estimation of the modified variance on the pathway within the samples from levels(phenotypes)[2].

$zscore

zscore for the modified variance.

$pvalue

theoretical p-value for null E1 = E2. (Standard EVA).

$pvalueD12D1

theoretical p-value for null E1 = E12.

$pvalueD12D2

theoretical p-value for null E2 = E12.

$pvalueTotal

Bonferonni corrected p-value of the three p-values.

Author(s)

Bahman Afsari

See Also

GSReg.GeneSets.DIRAC,cor

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
### loading and pruning the pathways
library(GSBenchMark)
data(diracpathways)
### loading the data
data(leukemia_GSEA)


### removing genes which contain not a number.
if(sum(apply(is.nan(exprsdata),1,sum)>0))
  exprsdata = exprsdata[-which(apply(is.nan(exprsdata),1,sum)>0),];
 
### extracting gene names
genenames = rownames(exprsdata);

### DIRAC analysis
VarAnKendallV = GSReg.GeneSets.EVA(geneexpres=exprsdata,
 pathways=diracpathways, phenotypes=as.factor(phenotypes))
E1 = sapply(VarAnKendallV,function(x) x$E1);
E2 = sapply(VarAnKendallV,function(x) x$E2);
Kpvalues = sapply(VarAnKendallV,function(x) x$pvalue);

dysregulatedpathways = rbind(E1[which(Kpvalues<0.05)],
E2[which(Kpvalues<0.05)],Kpvalues[which(Kpvalues<0.05)]);
rownames(dysregulatedpathways)<-c("E1","E2","pvalues");
print(dysregulatedpathways)
plot(x=dysregulatedpathways["E1",],y=dysregulatedpathways["E2",],
xlim=range(dysregulatedpathways[1:2,]),ylim=range(dysregulatedpathways[1:2,]))
lines(x=c(min(dysregulatedpathways[1:2,]),max(dysregulatedpathways[1:2,])),
y=c(min(dysregulatedpathways[1:2,]),max(dysregulatedpathways[1:2,])),type="l")

afsari/GSReg documentation built on May 9, 2019, 3:39 p.m.