testGE: Test the differential expression of the the genes in a...

In geneSignatureFinder: A Gene-signatures finder tools

Description

Given the clustering of the samples in good and poor prognosis associated to the signature, for each gene in the signature the test for the null hypothesis of equality of the expression levels is performed. Additional statistics are provided.

Usage

testGE(aSignatureFinder, permutationReplications = 1000, 
  cpuCluster = NULL, stopCpuCluster = TRUE)

Arguments

aSignatureFinder	(structure) as results from the function signatureFinder().
permutationReplications	(integer) number of replications of the permutation test (default: 1000).
cpuCluster	structure as result from the NCPUS() function
stopCpuCluster	flag to control if the channel to the cpu-cluster has to be closed.

Details

The t-test for testing the differential expression of the genes in the signature is performed according to the procedure of Dudoit et al. (2002). The test statistics is the Welch's one and the null distribution is obtained through a permutation scheme.

Value

The function returns the same variable in the input aSignatureFinder structure and

groupMedian	real matrix with as many rows as length(aSignatureFinder$signature) and two columns containing the medians of each gene with respect to the good and poor prognosis group
medianAbsDifference	a list of real with as many elements as length(aSignatureFinder$signature) where each entry is the absolute difference of the medians computed in each group
groupMean	real matrix with as many rows as length(signature$signature) and two columns containing the means of each gene with respect to the good and poor prognosis group
meanAbsDifference	a list of real with as many elements as length(aSignatureFinder$signature) where each entry is the absolute difference of the means computed in each group
meanDifferenceTValue	a list of real with as many elements as length(aSignatureFinder$signature) where each entry is the value of the test statistic
meanDifferencePValue	a list of real with as many elements as length(aSignatureFinder$signature) where each entry is the p-value of the test statistic

Author(s)

Stefano M. Pagnotta and Michele Ceccarelli

References

Dudoit S. et al.: Statistical methods for identifying differentially expressed genes in replicated cDNA microarray experiments Statistica Sinica, 12, pp. 111–139, 2002.

See Also

signatureFinder,

Examples

# find the signature starting from the gene SELP for  the Non Small Cell Lung Cancer 
# set the working data 
data(geNSCLC)
geData <- geNSCLC
data(stNSCLC)
stData <- stNSCLC
# set the dimension of the cpu's cluster 
# (use a value different from 2 depending on the number of cpu avalilable)
aMakeCluster <- makeCluster(2)
# set the starting gene to SELP
geneSeed <- which(colnames(geData) == "SELP")
# run ...
ans <- signatureFinder(geneSeed, logFilePrefix = "test", 
		cpuCluster = aMakeCluster, stopCpuCluster = FALSE)
ans
ans <- testGE(ans, cpuCluster = aMakeCluster)
ans$groupMean
ans$meanDifferencePValue
#####################
#library(gplots)
#barplot2(t(ans$groupMean), beside = TRUE,
#                   main = paste("Signature starting from:", ans$startingSignature),
#                   legend = paste(colnames(ans$groupMedian), "prognosis group"))

geneSignatureFinder documentation built on May 2, 2019, 2:32 p.m.