puma (version 3.12.0)

pumaNormalize: Normalize an ExpressionSet

Description

This is used to apply a scaling normalization to set of arrays. This normalization can be at the array scale (thus giving all arrays the same mean or median), or at the probeset scale (thus giving all probesets the same mean or median).

It is generally recommended that the default option (median array scaling) is used after running mmgmos and before running pumaComb and/or pumaDE. There are however, situations where this might not be the recommended, for example in time series experiments where it is expected than there will be general up-regulation or down-regulation in overall gene expression levels between time points.

Usage

pumaNormalize(
	eset
,	arrayScale = c("median", "none", "mean", "meanlog")
,	probesetScale = c("none", "mean", "median")
,	probesetNormalisation = NULL
,	replicates = list(1:dim(exprs(eset))[2])
)

Arguments

eset

An object of class ExpressionSet.

arrayScale

A method of scale normalisation at the array level.

probesetScale

A method of scale normalisation at the probe set level.

probesetNormalisation

If not NULL normalises the expression levels to have zero mean and adjusts the variance of the gene expression according to the zero-centered normalisation.

replicates

List of integer vectors indicating which arrays are replicates.

Value

An object of class ExpressionSet holding the normalised data.

See Also

Methods mmgmos, pumaComb and pumaDE

Examples

	#	Next 4 lines commented out to save time in package checks, and saved version used
    # if (require(affydata)) {
	#	data(Dilution)
	#	eset_mmgmos <- mmgmos(Dilution)
	# }
	data(eset_mmgmos)
	apply(exprs(eset_mmgmos),2,median)
	eset_mmgmos_normd <- pumaNormalize(eset_mmgmos)
	apply(exprs(eset_mmgmos_normd),2,median)