ELBOW (version 1.8.0)

get_elbow_limma: Calculates fold values from an MArrayLM object.

Description

Performs the ELBOW fold change test on an MArrayLM object. This is a wrapper class to help integrate the ELBOW method into Bioconductor. followed tutorial from: http://bioinformatics.knowledgeblog.org/2011/06/20/analysing-microarray-data-in-bioconductor/

Usage

get_elbow_limma(marraylm, columns = NULL)

Arguments

marraylm

is the MArrayLM object to analyze.

columns

is the list of sample columns to obtain the elbow fold cut-off values for. This can be specified as a vector or a single value.

Value

a matrix specified as follows

• columns --- (1) “up_limit”, the upper ELBOW fold-change cut-off value; (2) “low_limit”, the lower ELBOW fold-change cut-off value
• rows --- one row per sample, specified by the parameter “columns.”

Details

Some of the code in this method is based on how toptable accesses the MArrayLM object to read fold values. Therefore, any MArrayLM object which works with the toptable method should also work with this method.

Examples

########
		# LOAD DATA INTO LIMMA
		########
		library("limma")

		# load a filtered expression set into R
		# NOTE: see the vignette for instructions on preparing
		#       a filtered dataset with your own data.
		data(GSE20986_eset_exprs, package="ELBOW")
		data(GSE20986_design, package="ELBOW")

		# fit the linear model to the filtered expression set
		fit <- lmFit(GSE20986_eset_exprs, GSE20986_design)

		# set up a contrast matrix to compare tissues v cell line
		contrast.matrix <- makeContrasts(huvec_choroid = huvec - choroid, huvec_retina = huvec - retina, huvec_iris <- huvec - iris, levels=GSE20986_design)

		# Now the contrast matrix is combined with the per-probeset linear model fit.
		huvec_fits <- contrasts.fit(fit, contrast.matrix)
		huvec_ebFit <- eBayes(huvec_fits)

		########
		# GET THE ELBOW LIMIT (this function)
		########
		get_elbow_limma(huvec_ebFit)