HCA on PCA scores from a Spectra Object

Description

A wrapper which performs HCA on the scores from a PCA of a Spectra object, color-coding the results as specified in the object. Many methods for computing the clusters and distances are available.

Usage

1
2
3
hcaScores(spectra, pca, scores = c(1:5), c.method = "complete",
	d.method = "euclidean", use.sym = FALSE,
        leg.loc = "topright", ...)

Arguments

spectra

An object of S3 class Spectra.

pca

An object of class prcomp, modified to include a list element called $method, a character string describing the pre-processing carried out and the type of PCA performed (it appears on the plot). This is automatically provided if ChemoSpec functions c_pcaSpectra or r_pcaSpectra were used to create pca.

scores

A vector of integers specifying which scores to use for the HCA.

c.method

A character string describing the clustering method; must be acceptable to hclust.

d.method

A character string describing the distance calculation method; must be acceptable as a method in rowDist.

use.sym

A logical; if true, use no color and use lower-case letters to indicate group membership.

leg.loc

Character; if "none" no legend will be drawn. Otherwise, any string acceptable to legend.

...

Additional parameters to be passed to the plotting functions.

Value

A list, containing an object of class hclust and an object of class dendrogram. The side effect is a plot.

Author(s)

Bryan A. Hanson, DePauw University. hanson@depauw.edu

References

https://github.com/bryanhanson/ChemoSpec

See Also

hclust for the underlying function. See hcaSpectra for HCA of the entire data set stored in the Spectra object. plotHCA for the function that actually does the plotting.

Examples

1
2
3
4
data(SrE.IR)
results <- c_pcaSpectra(SrE.IR, choice = "noscale")
myt <- expression(bolditalic(Serenoa)~bolditalic(repens)~bold(IR~Spectra))
res <- hcaScores(SrE.IR,  results, scores = c(1:5), main = myt)

Want to suggest features or report bugs for rdrr.io? Use the GitHub issue tracker.