extendtree: Hierarchical clustering with amap::Dist and draw dendrogram...

In shkonishi/rsko: This is my personal package of R functions

Hierarchical clustering with amap::Dist and draw dendrogram with dendextend

Description

This functions returns the list consit of hclust object and dendrogram object. dendrogram is coloured by factor

Usage

extendtree(dat, extend, distm, clm, lab, gp, vcol, cexlab, blwd, ...)

Arguments

dat	data frame or matrix
extend	use with dendextend or not
distm	distance measure one of the amap::Dist methods "euclidean", "maximum", "manhattan", "canberra", "binary", "pearson", "abspearson", "correlation", "abscorrelation", "spearman" or "kendall". The default is "spearman".
clm	one of the hclust methods "ward.D", "ward.D2", "single", "complete", "average" (= UPGMA), "mcquitty"(= WPGMA), "median"(= WPGMC) or "centroid"(= UPGMC). The default is "ward.D2" .
lab	Leaf labels
gp	Group of labels as factor
vcol	A vector of colours corresponding to group of leaves
cexlab	cex of leaf label. The default is 0.8
blwd	line width of edges. The default is 1
...	arbitary option, if not using dendextend. E.g. nodePar = list(pch = NA, lab.cex = 0.8), edgePar = list(lwd = 2), leaflab ="none"

Value

Character vector, length of n

Examples

## Not run:  # minimum usage
res1 <- rsko::extendtree(dat = mtcars, extend = FALSE, lab = rownames(mtcars))
res1[[1]]
plot(res1[[2]])

# using dist object from other distance measure
cos.mat <- rsko::siml_mat(dat = t(iris[,1:4]), method = "cosine", type = "distance")
dist.cos <- as.dist(cos.mat)
res2 <-  rsko::extendtree(dat = iris[,1:4], extend = TRUE, distm = dist.cos, clm = "average",
    lab = as.character(iris$Species), gp = iris$Species, vcol = 1:3, cexlab = 0.5, blwd = 1)
plot(res2[[2]])

# using dendextend with several 'set' options
res3 <- rsko::extendtree(dat = iris[,1:4], extend = TRUE, distm = "euclidean", clm = "average",
    lab = as.character(iris$Species), gp = iris$Species, vcol = 1:3, cexlab = 0.5, blwd = 1.5)
plot(res3[[2]])

## End(Not run)

shkonishi/rsko documentation built on Feb. 21, 2023, 5:12 a.m.