affiXcanPca: Perform a PCA on each experiment found in...

In alussana/AffiXcan: A Functional Approach To Impute Genetically Regulated Expression

Description

Perform a PCA on each experiment found in MultiAssayExperiment objects

Usage

affiXcanPca(
  tbaPaths,
  varExplained = 80,
  scale = TRUE,
  regionsCount,
  BPPARAM = bpparam(),
  trainingSamples
)

Arguments

tbaPaths	A vector of strings, which are the paths to MultiAssayExperiment RDS files containing the tba values
varExplained	An integer between 0 and 100; varExplained=80 means that the principal components selected to fit the models must explain at least 80 percent of variation of TBA values; default is 80
scale	A logical; if scale=FALSE the TBA values will be only centered, not scaled before performing PCA; default is TRUE
regionsCount	An integer, that is the summation of length(assays()) of every MultiAssayExperiment RDS object indicated in the param tbaPaths; it is the returning value from overlookRegions()
BPPARAM	A BiocParallelParam object. Default is bpparam(). For details on BiocParallelParam virtual base class see browseVignettes("BiocParallel")
trainingSamples	A vector of strings. The identifiers (e.g. row names of MultiAssayExperiment objects from tbaPaths) of the samples that have to be considered in the training phase, and not used for the cross-validation

Value

pca: A list containing lists named as the MultiAssayExperiment::experiments() found in the MultiAssayExperiment objects listed in the param tbaPaths. Each of these lists contain two objects:

• eigenvectors: A matrix containing eigenvectors for those principal components selected according to the param varExplained

• pcs: A matrix containing the principal components values selected according to the param varExplained

• eigenvalues: A vector containing eigenvalues for those principal components selected according to the param varExplained

Examples

if (interactive()) {
data(exprMatrix)

tbaPaths <- system.file("extdata","training.tba.toydata.rds",
package="AffiXcan")
regionsCount <- overlookRegions(tbaPaths)

sampleNames <- colnames(exprMatrix)
nSamples <- length(sampleNames)
sampGroups <- subsetKFold(k=3, n=nSamples)
for (i in seq(length(sampGroups))) {
     sampGroups[[i]] <- colnames(exprMatrix)[sampGroups[[i]]]
}

testingSamples <- sampGroups[[1]]
trainingSamples <- sampleNames[!sampleNames %in% testingSamples]

pca <- affiXcanPca(tbaPaths=tbaPaths, varExplained=80, scale=TRUE,
regionsCount=regionsCount, trainingSamples=trainingSamples)
}

alussana/AffiXcan documentation built on Oct. 20, 2020, 12:08 a.m.