computePcaSample: Principal Components Analysis

In RclusTool: Graphical Toolbox for Clustering and Classification of Data Frames

Principal Components Analysis

Description

Perform Principal Components Analysis, dealing with the number of dimensions, automatically or not.

Usage

computePcaSample(
  data.sample,
  pca.nb.dims = 0,
  selected.var = NULL,
  echo = FALSE,
  prcomp.options = list(center = TRUE, scale = TRUE),
  pca.variance.cum.min = 0.95
)

Arguments

data.sample	list containing features, profiles and clustering results.
pca.nb.dims	number of dimensions to keep after PCA. If pca.nb.dims=0 (default), this number is automatically computed.
selected.var	vector of features names to consider for the PCA.
echo	boolean: if FALSE (default), no description printed in the console.
prcomp.options	list of default parameters values for the function prcomp.
pca.variance.cum.min	minimal cumulative variance to retain in PCA.

Details

computePcaSample performs Principal Components Analysis, dealing with the number of dimensions, automatically or not

Value

features list containing the results of PCA, returned by prcomp.

See Also

computeSpectralEmbeddingSample

Examples

dat <- rbind(matrix(rnorm(100, mean = 0, sd = 0.3), ncol = 2), 
             matrix(rnorm(100, mean = 2, sd = 0.3), ncol = 2), 
             matrix(rnorm(100, mean = 4, sd = 0.3), ncol = 2))
tf <- tempfile()
write.table(dat, tf, sep=",", dec=".")
x <- importSample(file.features=tf)

res <- computePcaSample(x, echo = TRUE)

plot(res$pca_full$x[,1], res$pca_full$x[,2], main="PCA", xlab="PC1", ylab="PC2")

RclusTool documentation built on Aug. 29, 2022, 9:07 a.m.