BoutrosLab.plotting.general (version 5.9.2)

create.dendrogram: Generate a dendrogram

Description

Takes a matrix and creates a row-wise or column-wise dendrogram

Usage

create.dendrogram(
	x,
	clustering.method = 'diana',
	cluster.dimension = 'col',
	distance.method = 'correlation',
	cor.method = 'pearson',
	force.clustering = FALSE,
	same.as.matrix = FALSE
	);

Arguments

x

A matrix that is used to create the dendrogram

clustering.method

Method used to cluster the records (can not be none). Accepts all agglomerative clustering methods available in hclust, plus “diana” (which is divisive).

cluster.dimension

Should clustering be performed on the rows or columns of x?

distance.method

Method name of the distance measure to be used for clustering. Defaults to “correlation”. Other supported methods are same as in ?dist. Also supports “jaccard” which is useful for clustering categorical variables.

cor.method

The method used for calculating correlation. Defaults to “pearson”

force.clustering

Binary to over-ride the control that prevents clustering of too-large matrices

same.as.matrix

Prevents the flipping of the matrix that the function normally does

Value

Returns an object of the dendrogram class corresponding to the row-wise or column-wise dendrogram for x

Examples

# NOT RUN {
# create temp data
x <- outer(-5:5, -5:5, '*') + matrix(nrow = 11, ncol = 11, data = runif(11 * 11));
colnames(x) <- paste('col', 1:11, sep = '-');
rownames(x) <- paste('row', 1:11, sep = '-');

# example of generating a column-wise dendrogram using default values
create.dendrogram(
    x = x
    );

# example of generating a column-wise dendrogram using different distance and clustering methods
create.dendrogram(
    x = x,
    clustering.method = 'median',
    cluster.dimension = 'cols',
    distance.method = 'euclidean'
    );

# generate row-wise dendrogram using default distance and clustering methods
create.dendrogram(
    x = x,
    cluster.dimension = 'row'
    );

# generate row-wise dendrogram using different distance and clustering methods
create.dendrogram(
    x = x,
    clustering.method = 'ward',
    cluster.dimension = 'rows',
    distance.method = 'manhattan'
    );
# }

Run the code above in your browser using DataCamp Workspace