plot_contribution_heatmap: Plot signature contribution heatmap

In CuppenResearch/MutationalPatterns: Comprehensive genome-wide analysis of mutational processes

Plot signature contribution heatmap

Description

Plot relative contribution of signatures in a heatmap

Usage

plot_contribution_heatmap(
  contribution,
  sig_order = NA,
  sample_order = NA,
  cluster_samples = TRUE,
  cluster_sigs = FALSE,
  method = "complete",
  plot_values = FALSE
)

Arguments

contribution	Signature contribution matrix
sig_order	Character vector with the desired order of the signature names for plotting. Optional.
sample_order	Character vector with the desired order of the sample names for plotting. Optional.
cluster_samples	Hierarchically cluster samples based on euclidean distance. Default = T.
cluster_sigs	Hierarchically cluster sigs based on euclidean distance. Default = T.
method	The agglomeration method to be used for hierarchical clustering. This should be one of "ward.D", "ward.D2", "single", "complete", "average" (= UPGMA), "mcquitty" (= WPGMA), "median" (= WPGMC) or "centroid" (= UPGMC). Default = "complete".
plot_values	Plot relative contribution values in heatmap. Default = F.

Value

Heatmap with relative contribution of each signature for each sample

See Also

extract_signatures, mut_matrix, plot_contribution, plot_cosine_heatmap

Examples

## Extracting signatures can be computationally intensive, so
## we use pre-computed data generated with the following command:
# nmf_res <- extract_signatures(mut_mat, rank = 2)

nmf_res <- readRDS(system.file("states/nmf_res_data.rds",
  package = "MutationalPatterns"
))

## Set signature names as row names in the contribution matrix
rownames(nmf_res$contribution) <- c("Signature A", "Signature B")

## Plot with clustering.
plot_contribution_heatmap(nmf_res$contribution, cluster_samples = TRUE, cluster_sigs = TRUE)

## Define signature and sample order for plotting. If you have a mutation or signature
## matrix, then this can be done like in the example of 'plot_cosine_heatmap()'
sig_order <- c("Signature B", "Signature A")
sample_order <- c(
  "colon1", "colon2", "colon3", "intestine1", "intestine2",
  "intestine3", "liver3", "liver2", "liver1"
)
plot_contribution_heatmap(nmf_res$contribution,
  cluster_samples = FALSE,
  sig_order = sig_order, sample_order = sample_order
)

## It's also possible to create a contribution heatmap with text values
output_text <- plot_contribution_heatmap(nmf_res$contribution, plot_values = TRUE)

## This function can also be used on the result of a signature refitting analysis.
## Here we load a existing result as an example.
snv_refit <- readRDS(system.file("states/strict_snv_refit.rds",
  package = "MutationalPatterns"
))
plot_contribution_heatmap(snv_refit$contribution, cluster_samples = TRUE, cluster_sigs = TRUE)

CuppenResearch/MutationalPatterns documentation built on Nov. 23, 2022, 4:13 a.m.