buildIcaSet: This function builds an object of class 'IcaSet'.
In bitona/MineICA: Analysis of an ICA decomposition obtained on genomics data

buildIcaSet

R Documentation

This function builds an object of class `IcaSet`.

Description

This function builds an object of class IcaSet.

Usage

  buildIcaSet(params, A, S, dat, pData = new("data.frame"),
    fData = new("data.frame"), witGenes = new("character"),
    compNames = new("character"),
    refSamples = new("character"),
    annotation = new("character"),
    chipManu = new("character"),
    chipVersion = new("character"), alreadyAnnot = FALSE,
    typeID = c(geneID_annotation = "SYMBOL", geneID_biomart = "hgnc_symbol", featureID_biomart = ""),
    runAnnot = TRUE, organism = "Human",
    mart = new("Mart"))

Arguments

`params`	An object of class `MineICAParams` containing the parameters of the analysis
`A`	The mixing matrix of the ICA decomposition (of dimension samples x components).
`S`	The source matrix of the ICA decomposition (of dimension features x components).
`dat`	The data matrix the ICA was applied to (of dimension features x samples).
`pData`	Phenotype data, a data.frame which contains the sample informations of dimension samples x annotations.
`fData`	Feature data, a data.frame which contrains the feature descriptions of dimensions features x annotations.
`witGenes`	A vector of witness genes. They are representative of the expression behavior of the contributing genes of each component. If missing or NULL, they will be automatically attributed using function `selectWitnessGenes`.
`compNames`	A vector of component labels.
`refSamples`	A vector of reference sample IDs (e.g the "normal" samples).
`annotation`	An annotation package (e.g a ".db" package specific to the microarray used to generate `dat`)
`chipManu`	If microarray data, the manufacturer: either 'affymetrix' or 'illumina'.
`chipVersion`	For illumina microarrays: the version of the microarray.
`alreadyAnnot`	TRUE if the feature IDs contained in the row names of `dat` and `S` already correspond to the final level of annotation (e.g if they are already gene IDs). In that case, no annotation is performed.
`typeID`	A character vector specifying the annotation IDs, it includes three elements : geneID_annotation the IDs from the package to be used to annotate the features into genes. It will be used to fill the attributes `datByGene` and `SByGene` of the `icaSet`. It must match one of the objects the corresponding package supports (you can access the list of objects by typing ls("package:packagename")). If no annotation package is provided, this element is not useful. geneID_biomart the type of gene IDs, as available in `listFilters(mart)`; where mart is specified as described in `useMart`. If you have directly built the IcaSet at the gene level (i.e if no annotation package is used), `featureID_biomart` and `geneID_biomart` will be identical. featureID_biomart the type of feature IDs, as available in `listFilters(mart)`; where `mart` is specified as described in function `useMart`. Not useful if you work at the gene level.
`runAnnot`	If TRUE, `icaSet` is annotated with function `annotInGene`.
`organism`	The organism the data correspond to.
`mart`	The mart object (database and dataset) used for annotation, see function `useMart` of package `biomaRt`

Value

An object of class IcaSet

Author(s)

Anne Biton

Examples

dat <- data.frame(matrix(rnorm(10000),ncol=10,nrow=1000))
rownames(dat) <- paste("g", 1:1000, sep="")
colnames(dat) <- paste("s", 1:10, sep="")

## build a data.frame containing sample annotations
annot <- data.frame(type=c(rep("a",5),rep("b",5)))
rownames(annot) <- colnames(dat)

## run ICA
resJade <- runICA(X=dat, nbComp=3, method = "JADE")

## build params
params <- buildMineICAParams(resPath="toy/")

## build IcaSet object
icaSettoy <- buildIcaSet(params=params, A=data.frame(resJade$A), S=data.frame(resJade$S),
                         dat=dat, pData=annot, alreadyAnnot=TRUE)
params <- icaSettoy$params
icaSettoy <- icaSettoy$icaSet

## Not run: 
## load data
library(breastCancerMAINZ)
data(mainz)

## run ICA
resJade <- runICA(X=dataMainz, nbComp=10, method = "JADE", maxit=10000)

## build params
params <- buildMineICAParams(resPath="mainz/")

## build IcaSet object

# fill typeID, Mainz data originate from affymetrix HG-U133a  microarray and are indexed by probe sets
# we want to annotate the probe sets into Gene Symbols
typeIDmainz <-  c(geneID_annotation="SYMBOL", geneID_biomart="hgnc_symbol", featureID_biomart="affy_hg_u133a")

icaSetMainz <- buildIcaSet(params=params, A=data.frame(resJade$A), S=data.frame(resJade$S),
                             dat=exprs(mainz), pData=pData(mainz),
                             annotation="hgu133a.db", typeID= c(geneID_annotation = "SYMBOL",
                             geneID_biomart = "hgnc_symbol", featureID_biomart = "affy_hg_u133a"),
                             chipManu = "affymetrix", runAnnot=TRUE,
                             mart=useMart(biomart="ensembl", dataset="hsapiens_gene_ensembl"))

## End(Not run)

bitona/MineICA documentation built on April 23, 2023, 1:41 p.m.