Skip to contents

Plot a delta-signal modification heatmap.

Source: R/plots.R
plot_mod_heatmap.Rd

Create a diverging heatmap of modification signal changes (e.g., mutant minus wild-type) across tRNA families and Sprinzl positions. Rows can be optionally clustered using Ward's D2 hierarchical clustering.

Usage

plot_mod_heatmap(
  data,
  value_col = "value",
  ref_col = "ref",
  cluster = TRUE,
  color_limits = c(-0.25, 0.25),
  color_low = "#0072B2",
  color_high = "#D55E00",
  na_value = "gray80",
  square = TRUE,
  label_col = NULL,
  label_min = 0.05,
  label_size = 2.5,
  highlight_col = NULL,
  highlight_size = 0.4,
  highlight_offset = c(-0.35, 0.35),
  cluster_threshold = NULL,
  group_col = NULL,
  divider_linewidth = 0.8,
  fill_name = waiver(),
  fill_breaks = waiver(),
  caption = NULL
)

Arguments

data

A data frame with at least three columns: one for tRNA family/reference (y-axis), one for Sprinzl position labels (x-axis), and one for the fill value.

value_col

Column name (string) for fill values. Default "value".

ref_col

Column name (string) for tRNA families (y-axis). Default "ref".

cluster

Logical; cluster rows with Ward's D2? Default TRUE.

color_limits

Numeric vector of length 2 giving symmetric limits for the color scale. Default c(-0.25, 0.25).

color_low

Color for negative values. Default "#0072B2" (blue).

color_high

Color for positive values. Default "#D55E00" (red).

na_value

Color for missing positions. Default "gray80".

square

Logical; use coord_fixed(ratio = 1)? Default TRUE.

label_col

Column name (string) with text labels to overlay on tiles (e.g., nucleotide letters). Default NULL (no labels).

label_min

Minimum abs(value) to show a label. Default 0.05.

label_size

Font size for tile labels. Default 2.5.

highlight_col

Column name (string) of a logical column; TRUE cells get a dot overlay. Default NULL (no dots).

highlight_size

Dot size for highlighted cells. Default 0.8.

highlight_offset

Numeric vector of length 2 giving x/y offsets from tile center for highlight dots. Default c(-0.35, 0.35).

cluster_threshold

Numeric threshold for noise filtering during clustering. When non-NULL, only positions where any row has abs(value) > cluster_threshold are used to build the distance matrix. Falls back to all positions if nothing passes. Default NULL.

group_col

Column name (string) for group-aware clustering. When provided, rows are clustered within each group and horizontal divider lines separate groups. Default NULL.

divider_linewidth

Line width for group dividers. Default 0.8.

fill_name

Legend title for the fill scale. Default waiver() (ggplot2 default).

fill_breaks

Numeric vector of legend breaks for the fill scale. Default waiver() (ggplot2 default).

caption

Explanatory text displayed below the plot. Default NULL.

Value

A ggplot object.

Examples

df <- tidyr::expand_grid(
  ref = paste0("tRNA-", c("Ala", "Gly", "Ser")),
  sprinzl_label = as.character(1:10)
)
df$value <- rnorm(nrow(df), sd = 0.1)
plot_mod_heatmap(df)