rmatrain: RMA normalization with "documentation by value" (Kostka &...

In bapred: Batch Effect Removal and Addon Normalization (in Phenotype Prediction using Gene Data)

RMA normalization with “documentation by value” (Kostka & Spang, 2008)

Description

Performs RMA normalization and returns the normalized dataset together with information necessary for addon RMA normalization (Kostka & Spang, 2008) using rmaaddon.

Usage

rmatrain(affybatchtrain)

Arguments

affybatchtrain

object of class AffyBatch. Affymetrix GeneChip probe level data.

Details

This function uses code from the off-CRAN package docval, version 1.0.

Value

rmatrain returns an object of class rmatrain. An object of class "rmatrain" is a list containing the following components:

xnorm	matrix of RMA normalized (training) data. Observations in rows, variables in columns.
rmadoc, sumdoc.rma, nfeature	information necessary for addon RMA normalization

Author(s)

Roman Hornung

References

Kostka, D., Spang, R. (2008). Microarray based diagnosis profits from better documentation of gene expression signatures. PLoS Computational Biology 4(2):e22, <doi: 10.1371/journal.pcbi.0040022>.

Examples

## Not run: 

# Read in example data from ArrayExpress-webpage:

library("ArrayExpress")

expFiles <- getAE("E-GEOD-62837", path = tempdir(), type = "raw")

rawfiles <- file.path(tempdir(), expFiles$rawFiles)

library("affy")
# Training data:
affybatchtrain <- ReadAffy(filenames=rawfiles[1:3])

try(file.remove(file.path(tempdir(), expFiles$rawFiles)))
try(file.remove(file.path(tempdir(), expFiles$processedFiles)))
try(file.remove(file.path(tempdir(), expFiles$sdrf)))
try(file.remove(file.path(tempdir(), expFiles$idf)))
try(file.remove(file.path(tempdir(), expFiles$adf)))
try(file.remove(file.path(tempdir(), expFiles$rawArchive)))
try(file.remove(file.path(tempdir(), expFiles$processedArchive)))


# RMA normalization with documentation by value:
rmaparams <- rmatrain(affybatchtrain)
Xtrainnorm <- rmaparams$xnorm
dim(Xtrainnorm)

## End(Not run)

bapred documentation built on June 22, 2022, 9:08 a.m.