simpleaffy (version 2.48.0)

pairwise.comparison: Compute pairwise comparison statistics between two experimental groups

Description

Generate fold changes, t-tests and means for a pair of experimental groups

Usage

pairwise.comparison(x,group,members=NULL,spots=NULL,a.order=NULL,b.order=NULL,method="unlogged",logged=TRUE)

Arguments

x

an ExpressionSet object.

group

column in pData(x).

members

labels in group.

spots

unnormalised AffyBatch data for this experiment - if included, results in PMA calls and detection p-values being generated

a.order

For a comparison with matched pairs, the ordering of the first group of replicates

b.order

For a comparison with matched pairs, the ordering of the second group of replicates

method

What method should be used to calculate the average for the fold-change - can be either "logged","unlogged","median"

logged

Whether the input data is logged or not

Value

A Pairwise comparison object.

Details

Given an ExpressionSet object, generate quick stats for pairwise comparisons between a pair of experimental groups. If a.order and b.order are specified then a paired sample t-test will be conducted between the groups, with the arrays in each group sorted according to the ordering specified. By default, the function assumes that the expression values are logged (this can be changed with the parameter "logged"). The fold-changes are computed from the average values across replicates. Unless you specify otherwise, this is done using the mean of the unlogged values (i.e. logged data is first unlogged, the mean calculated, and the result re-logged). The parameter "method", allows the mean of the logged values or their median to be used instead. T-tests are always computed with the logged data.

References

http://bioinformatics.picr.man.ac.uk/

Examples

  ## Not run: 
#     pc <- pairwise.comparison(eset.rma,"group",c("A","P")) 
#   ## End(Not run)