plw (version 1.32.0)

logitT: logit-t and t-test by row

Description

Functions for the logit-t test (Lemon et al. 2003) and the ordinary t-test computed for each row of an matrix.

Usage

logitTTransform(pm)
logitTStat(affy.batch,group)
studenttTTest(x, group)

Arguments

pm

A matrix of Pm intensities

affy.batch

An AffyBatch object

group

A group indicator vector, should have values 1 and 2 only.

x

A matrix

Value

logitTTransform returns a matrixlogitTStat returns a vector with the logit-t statistic for each probe set.studenttTTest returns a vector with t-statistic for each row of x.

Details

See the definition (R-code) of each function for details.

References

Lemon et al. (2003). A high performance test of differential gene expression for oligonucleotide arrays. Genome Biol. 2003; 4(10):R67

Examples

# ------------------------------------------
# Example analyzing the 6 arrays in the 
# AffySpikeU95Subset data set

# Loading the data
data(AffySpikeU95Subset)

# Vector with groups assignment 
group<-factor(rep(1:2,each=3))

# logit-T statistic
logitT<-logitTStat(AffySpikeU95Subset,
                   as.numeric(group))

# Computing RMA expression index
data.rma<-exprs(rma(AffySpikeU95Subset))


# Ordinary t-test by row/gene
studentT<-studenttTTest(data.rma, as.numeric(group))

# Comparing genes ranked top-20
logitTTop20  <- rank(-abs(logitT)) < 21
studentTTop20<- rank(-abs(studentT)) < 21
table(logitTTop20,studentTTop20)