cluster_matrix {textshape}R Documentation

Reorder a Matrix Based on Hierarchical Clustering

Description

Reorder matrix rows, columns, or both via hierarchical clustering.

Usage

cluster_matrix(x, dim = "both", method = "ward.D2", ...)

Arguments

x

A matrix.

dim

The dimension to reorder (cluster); must be set to "row", "col", or "both".

method

The agglomeration method to be used (see hclust).

...

ignored.

Value

Returns a reordered matrix.

See Also

hclust

Examples

cluster_matrix(mtcars)
cluster_matrix(mtcars, dim = 'row')
cluster_matrix(mtcars, dim = 'col')

## Not run: 
if (!require("pacman")) install.packages("pacman")
pacman::p_load(tidyverse, viridis, gridExtra)

## plot heatmap w/o clustering
wo <- mtcars %>%
    cor() %>%
    tidy_matrix('car', 'var') %>%
    ggplot(aes(var, car, fill = value)) +
         geom_tile() +
         scale_fill_viridis(name = expression(r[xy])) +
         theme(
             axis.text.y = element_text(size = 8)   ,
             axis.text.x = element_text(
                 size = 8, 
                 hjust = 1, 
                 vjust = 1, 
                 angle = 45
             ),   
             legend.position = 'bottom',
             legend.key.height = grid::unit(.1, 'cm'),
             legend.key.width = grid::unit(.5, 'cm')
         ) +
         labs(subtitle = "With Out Clustering")

## plot heatmap w clustering
w <- mtcars %>%
    cor() %>%
    cluster_matrix() %>%
    tidy_matrix('car', 'var') %>%
    mutate(
        var = factor(var, levels = unique(var)),
        car = factor(car, levels = unique(car))        
    ) %>%
    group_by(var) %>%
    ggplot(aes(var, car, fill = value)) +
         geom_tile() +
         scale_fill_viridis(name = expression(r[xy])) +
         theme(
             axis.text.y = element_text(size = 8)   ,
             axis.text.x = element_text(
                 size = 8, 
                 hjust = 1, 
                 vjust = 1, 
                 angle = 45
             ),   
             legend.position = 'bottom',
             legend.key.height = grid::unit(.1, 'cm'),
             legend.key.width = grid::unit(.5, 'cm')               
         ) +
         labs(subtitle = "With Clustering")

gridExtra::grid.arrange(wo, w, ncol = 2)

## End(Not run)

[Package textshape version 1.7.5 Index]