Skip to contents

Create Publication-Ready Oncoplot

Source: R/oncoplot.R
bb_oncoplot.Rd

Generates a waterfall-style oncoplot showing the mutation landscape across samples. Accepts either a simple data.frame with sample, gene, mutation_type columns or a MAF-format data.frame.

Usage

bb_oncoplot(
  data,
  genes = NULL,
  n_genes = 20L,
  sort_by = c("frequency", "cluster"),
  annotation_df = NULL,
  mutation_colors = NULL,
  show_pct = TRUE,
  show_barplot = TRUE,
  title = NULL,
  border_color = "white"
)

Arguments

data

A data.frame with columns sample, gene, mutation_type. Alternatively, a MAF-like data.frame with Hugo_Symbol, Tumor_Sample_Barcode, and Variant_Classification columns.

genes

Character vector. Specific genes to display. If NULL, the top n_genes most frequently mutated genes are shown.

n_genes

Integer. Number of top mutated genes if genes is NULL. Default 20.

sort_by

Character. How to sort samples: "frequency" (by mutation burden) or "cluster" (by co-occurrence clustering). Default "frequency".

annotation_df

A data.frame for sample annotations (bottom tracks). Rownames must be sample identifiers. Each column becomes an annotation track.

mutation_colors

Named character vector mapping mutation types to colors. If NULL, a default nature-style palette is used.

show_pct

Logical. Show mutation percentage per gene on the right margin. Default TRUE.

show_barplot

Logical. Show top (sample burden) and side (gene count) barplots. Default TRUE.

title

Character or NULL. Plot title.

border_color

Character or NULL. Color of tile borders. Default "white".

Value

A ggplot2::ggplot object (composed with patchwork if barplots are shown and patchwork is available).

Details

The oncoplot encodes the mutation landscape of a cohort as a tiled grid: columns = samples, rows = genes. Tiles are colored by mutation type. Multi-hit genes (multiple mutation types in one sample) are labeled "Multi_Hit".

When show_barplot = TRUE and the patchwork package is available, the plot is composed of three panels: top barplot (mutation burden per sample), main tile plot, and right barplot (mutations per gene). Without patchwork, only the main tile plot is returned.

Examples

set.seed(42)
mut_data <- data.frame(
  sample = sample(paste0("TCGA-", 1:30), 150, replace = TRUE),
  gene = sample(c("TP53","KRAS","PIK3CA","PTEN","APC",
                   "BRAF","EGFR","NRAS","CDKN2A","RB1"), 150, replace = TRUE),
  mutation_type = sample(c("Missense_Mutation","Nonsense_Mutation",
                           "Frame_Shift_Del","Splice_Site"), 150, replace = TRUE)
)
bb_oncoplot(mut_data, n_genes = 10)