Simple limma analysis on expression data with one contrast.

Share:

Description

This is a convenience function, which can be used to perform a limma analysis with one contrast. Subsequently results may be passed to birteLimma

Usage

1
limmaAnalysis(dat, design=NULL, contrast)

Arguments

dat

A matrix or ExpressionSet containing the expression values.

design

A design matrix. If no design matrix is provided, the method tries to infer a design matrix by taking column names as a factor. That means the design matrix is constructed according to ~0 + factor(colnames(dat).

contrast

Contrast for the linear model. Currently only one contrast is supported. If more contrasts are desired separate biRte analyses should be run for each of them.

Value

Returns a list containing the following entries:

pvalue.tab

Containing the result of the topTable function from limma.

lm.fit

Linear fit to the model, i.e. output of function 'eBayes'.

design

The design used.

contrast

The contrasts used.

Author(s)

Holger Froehlich

References

G. K. Smyth. Limma : Linear Models for Microarray Data. Bioinformatics, (2005):397-420.

Examples

1
# see birte.run