contingency_heatmap: Heatmap of contingency table

In jichunxie/CDI: Clustering Deviation Index (CDI)

Heatmap of contingency table

Description

If the benchmark cell type label set is available, we can also compare one candidate label set (e.g. the optimal clustering label set selected by CDI) with the benchmark cell type labels. Here we provide the heatmap of contingency table for comparison. Each row represents a cell type in the benchmark label set, and each column represents a cluster in the clustering label set. Each rectangle is color-scaled by the proportion of the cells in the given cluster coming from the benchmark types. Each column sums to 1.

Usage

contingency_heatmap(
  benchmark_label,
  candidate_label,
  proportion_size = 4,
  show_axis_names = TRUE,
  show_legend = TRUE,
  rename_candidate_clusters = FALSE,
  candidate_cluster_names = NULL
)

Arguments

benchmark_label	A vector of characters indicating the benchmark cell type label set of cells.
candidate_label	A vector of characters indicating the candidate clustering label set of cells.
proportion_size	A number indicating the label size of proportion values inside each rectangle. The label will not be shown if this parameter is set to be FALSE.
show_axis_names	A bool value indicating whether the axis names should be shown or not in the plot.
show_legend	A bool value indicating whether the legend of methods should be shown or not in the plot.
rename_candidate_clusters	A bool value indicating whether the candidate cluster names will be changed or not. If true, the candidate cluster names will be changed to the candidate_cluster_names.
candidate_cluster_names	A vector of characters indicating the candidate cluster names. The order of names will be the same as the order without customizing the candidate cluster names. For example, when rename_candidate_clusters is FALSE, and the output figure has x axis label (cluster1, cluster0). If rename_candidate_clusters is TRUE, and candidate_cluster_names is (c1, c2). That means cluster1 -> c1, and cluster0 -> c0.

Value

A ggplot object.

Examples

contingency_heatmap(
	benchmark_label = c(rep("type_b", 45), rep("type_a", 145), rep("type_c", 10)),
	candidate_label = paste0('cluster', c(rep(0,10), rep(1, 180), rep(2, 10))))
contingency_heatmap(
	benchmark_label = c(rep("type_b", 45), rep("type_a", 145), rep("type_c", 10)),
	candidate_label = paste0('cluster', c(rep(0,10), rep(1, 180), rep(2, 10))),
	rename_candidate_clusters = TRUE, 
	candidate_cluster_names = c('1', '2', '3'))

jichunxie/CDI documentation built on Dec. 6, 2023, 5:48 a.m.