Package {biopalette}

Title: Image-Inspired Color Palettes for Biomedical Visualization
Version: 0.2.2
Description: Provides a curated collection of image-inspired color palettes for biomedical visualization. The palettes are organized as qualitative, sequential, or diverging scales and include documented source context and intended use. The package provides functions to retrieve, inspect, preview, and apply these palettes in base R and 'ggplot2' graphics, together with utilities for working with palette definitions.
License: MIT + file LICENSE
URL: https://github.com/evanbio/biopalette, https://evanbio.github.io/biopalette/
BugReports: https://github.com/evanbio/biopalette/issues
Depends: R (≥ 4.1)
Imports: cli, ggplot2, jsonlite, scales
Encoding: UTF-8
Language: en-US
RoxygenNote: 7.3.2
Suggests: knitr, rmarkdown, testthat (≥ 3.0.0)
VignetteBuilder: knitr, rmarkdown
Config/testthat/edition: 3
NeedsCompilation: no
Packaged: 2026-08-22 15:45:05 UTC; 84241
Author: Yibin Zhou ORCID iD [aut, cre]
Maintainer: Yibin Zhou <evanzhou.bio@gmail.com>
Repository: CRAN
Date/Publication: 2026-09-02 12:10:02 UTC

biopalette: Image-Inspired Color Palettes for Biomedical Visualization

Description

The biopalette package provides image-inspired color palettes for biomedical visualization. Each palette is sourced from a real image — a film still, a scientific figure, or an artwork — and translated into a reproducible color system.

Details

Main features:

Author(s)

Maintainer: Yibin Zhou evanzhou.bio@gmail.com (ORCID)

See Also

Useful links:

Create and Save a Custom Color Palette

Description

Save a named color palette as a JSON file in a collection directory. The palette is usable immediately: point any reading function at the same palettes_dir. The JSON is written to a same-directory temporary file, validated, and then committed; a failed overwrite leaves the previous palette intact.

Usage

create_palette(
  name,
  type = c("sequential", "diverging", "qualitative"),
  colors,
  palettes_dir,
  overwrite = FALSE
)

Arguments

name

Character. Palette name (e.g., "blues").

type

Character. One of "sequential", "diverging", or "qualitative".

colors

Character vector of HEX color values (e.g., "#E64B35" or "#E64B35B2").

palettes_dir

Character. Directory to write the palette into. Required: there is deliberately no default, so a palette can never be written into the collection that ships with the package.

overwrite

Logical. If TRUE, overwrite existing palette file. Default: FALSE.

Value

Invisibly returns a list with path and info.

Examples

temp_dir <- file.path(tempdir(), "palettes")
create_palette("blues", "sequential", c("#deebf7", "#9ecae1", "#3182bd"),
  palettes_dir = temp_dir)
create_palette("qual_vivid", "qualitative", c("#E64B35", "#4DBBD5", "#00A087"),
  palettes_dir = temp_dir)

# Overwrite an existing palette explicitly
create_palette("blues", "sequential", c("#c6dbef", "#6baed6", "#2171b5"),
  palettes_dir = temp_dir, overwrite = TRUE)

unlink(temp_dir, recursive = TRUE)

Get a Color Palette

Description

Retrieve a named palette by name and type, returning a vector of HEX colors. Automatically checks for type mismatch and provides smart suggestions.

Usage

get_palette(name, type = NULL, n = NULL, reverse = FALSE, palettes_dir = NULL)

Arguments

name

Character. Name of the palette (e.g. "qual_vivid").

type

Character. One of "sequential", "diverging", "qualitative". If NULL, type is auto-detected.

n

Integer. Number of colors to return. If NULL, returns all colors. See What n means above. Default is NULL.

reverse

Logical. Reverse the palette before n is applied, so the two arguments stay independent: reverse hands back a different palette and n then selects from it. Default: FALSE.

palettes_dir

Character. Directory holding a palette collection (⁠sequential/⁠, ⁠diverging/⁠, ⁠qualitative/⁠ subdirectories of JSON files). If NULL, the palettes bundled with the package are used.

Value

Character vector of HEX color codes.

What n means

n is resolved according to what the palette's type says the colors are, because "give me 3 colors" means two different things:

qualitative

The colors are unordered categories, so n takes the first n of them. Asking for more than the palette holds is an error — there is no way to invent a category that the palette does not contain.

sequential, diverging

The colors are stops along a ramp, so n returns n steps spanning the whole ramp, interpolating as needed. Interpolation takes place in Lab colour space, matching the package's ggplot2 gradient scales. Any n works, above or below the number of stops. Taking the first n stops instead would silently hand back one end of the ramp — the light half of a sequential scale, or one arm of a diverging one.

n equal to the number of stops returns the palette untouched. When reverse = TRUE the palette is flipped first, so n selects from the reversed palette.

Examples

get_palette("gene_red", type = "qualitative")

# Qualitative: the first n categories
get_palette("walter_white2", type = "qualitative", n = 2)

# Sequential: n steps across the whole ramp, not the first n stops
get_palette("mitonuclear_blue")
get_palette("mitonuclear_blue", n = 3)

# Ramps can also be stretched beyond the stops they were drawn from
get_palette("walter_white", type = "diverging", n = 9)

# Flip a ramp end to end
get_palette("mitonuclear_blue", reverse = TRUE)

Convert HEX Colors to RGB

Description

Convert a character vector of HEX color codes to a data.frame with columns hex, r, g, b, and alpha.

Usage

hex2rgb(hex)

Arguments

hex

Character vector of HEX color codes (e.g. "#FF8000" or "#FF8000B2"). Both 6-digit and 8-digit (with alpha) codes are accepted. The ⁠#⁠ prefix is required. No NA values allowed.

Value

A data.frame with hex (character) and integer r, g, b, and alpha columns. Channels lie in ⁠[0, 255]⁠. alpha is NA_integer_ for a 6-digit input and the explicit alpha channel for an 8-digit input.

Examples

hex2rgb("#FF8000")
hex2rgb(c("#FF8000", "#00FF00"))

List Available Color Palettes

Description

Return a data.frame of all available palette metadata, optionally filtered by type.

Usage

list_palettes(type = NULL, sort = TRUE, palettes_dir = NULL)

Arguments

type

Palette type(s) to filter: "sequential", "diverging", "qualitative". Default NULL returns all.

sort

Whether to sort by type, n_color, name. Default: TRUE.

palettes_dir

Character. Directory holding a palette collection (⁠sequential/⁠, ⁠diverging/⁠, ⁠qualitative/⁠ subdirectories of JSON files). If NULL, the palettes bundled with the package are used.

Value

A data.frame with columns: name, type, n_color, colors.

Examples

list_palettes()
list_palettes(type = "qualitative")
list_palettes(type = c("sequential", "diverging"))

Description

Display palettes in a paged gallery format, returning a named list of ggplot objects.

Usage

palette_gallery(
  type = NULL,
  max_palettes = 30,
  max_row = 12,
  verbose = TRUE,
  palettes_dir = NULL
)

Arguments

type

Palette types to include: "sequential", "diverging", "qualitative". Default NULL returns all.

max_palettes

Number of palettes per page. Default: 30.

max_row

Max colors per row. Default: 12.

verbose

Whether to print progress info. Default: TRUE.

palettes_dir

Character. Directory holding a palette collection (⁠sequential/⁠, ⁠diverging/⁠, ⁠qualitative/⁠ subdirectories of JSON files). If NULL, the palettes bundled with the package are used.

Value

A named list of ggplot objects (one per page).

Examples


palette_gallery()
palette_gallery(type = "qualitative")
palette_gallery(type = c("sequential", "diverging"), max_palettes = 10)

Get Metadata for One Color Palette

Description

Return the runtime metadata for a single named palette. This is the one-palette counterpart to list_palettes() and uses the same lookup rules as get_palette().

Usage

palette_info(name, type = NULL, palettes_dir = NULL)

Arguments

name

Character. Palette name.

type

Character. One of "sequential", "diverging", or "qualitative". If NULL, the type is detected automatically.

palettes_dir

Character. Directory holding a palette collection. If NULL, the palettes bundled with the package are used.

Value

A one-row data.frame with columns name, type, n_color, and colors. The colors column is a list-column containing the complete HEX color vector.

Examples

palette_info("walter_white")
palette_info("babel", type = "qualitative")

Preview a Color Palette

Description

Visualize a palette using various plot styles.

Usage

preview_palette(
  name,
  type = NULL,
  n = NULL,
  reverse = FALSE,
  plot_type = c("bar", "pie", "point", "rect", "circle"),
  title = NULL,
  palettes_dir = NULL
)

Arguments

name

Character. Name of the palette.

type

Character. One of "sequential", "diverging", "qualitative". If NULL, auto-detected.

n

Integer. Number of colors to use. If NULL, uses all. Default: NULL.

reverse

Logical. Reverse the palette before n is applied. Passed straight through to get_palette(). Default: FALSE.

plot_type

Character. One of "bar", "pie", "point", "rect", "circle". Default: "bar".

title

Character. Plot title. If NULL, defaults to palette name.

palettes_dir

Character. Directory holding a palette collection (⁠sequential/⁠, ⁠diverging/⁠, ⁠qualitative/⁠ subdirectories of JSON files). If NULL, the palettes bundled with the package are used.

Details

"bar", "pie", "point" and "circle" are drawn with base graphics; "rect" is drawn with ggplot2. Either way the plot goes straight to the active device and nothing is returned — use palette_gallery() when you want plot objects you can modify, arrange or save.

Value

NULL, invisibly. Called for its plotting side effect — the return shape is the same for every plot_type.

See Also

palette_gallery(), which returns ggplot objects instead of drawing.

Examples


preview_palette("gene_red", plot_type = "bar")
preview_palette("walter_white", plot_type = "pie")
preview_palette("walter_white2", n = 2, plot_type = "circle")

Remove a Saved Palette JSON

Description

Remove a palette JSON file by name, searching across types if needed.

Usage

remove_palette(name, type = NULL, palettes_dir)

Arguments

name

Character. Palette name (without '.json' suffix).

type

Character. One of "sequential", "diverging", "qualitative". If NULL, searches all types.

palettes_dir

Character. Directory holding the palette collection. Required: there is deliberately no default, so the collection that ships with the package can never be removed from.

Value

Invisibly TRUE if removed successfully, FALSE otherwise.

Examples

temp_dir <- tempfile("biopalette-palettes-")
create_palette(
  "example_palette",
  "qualitative",
  c("#E64B35", "#4DBBD5", "#00A087"),
  palettes_dir = temp_dir
)

remove_palette("example_palette", palettes_dir = temp_dir)
unlink(temp_dir, recursive = TRUE)

Convert RGB Values to HEX Color Codes

Description

Convert RGB or RGBA values to HEX color codes. Accepts either a numeric vector or a data.frame, symmetrically with hex2rgb().

Usage

rgb2hex(rgb)

Arguments

rgb

A numeric vector of length 3 (r, g, b) or 4 (r, g, b, alpha), or a data.frame with r, g, b, and optional alpha columns. Matrices and arrays are not accepted. Values must lie in ⁠[0, 255]⁠ and are rounded to the nearest integer, with exact halves rounded up. In a data.frame, alpha = NA emits 6-digit HEX; a finite alpha value emits 8-digit HEX. Data frames must contain at least one row; unrelated extra columns are ignored.

Value

A character vector of uppercase 6- or 8-digit HEX color codes.

Examples

rgb2hex(c(255, 128, 0))
rgb2hex(c(255, 128, 0, 178))
rgb2hex(hex2rgb(c("#FF8000", "#00FF0080")))

Discrete ggplot2 Scales from a biopalette Palette

Description

Use a named biopalette palette for a discrete colour or fill mapping. The scale asks get_palette() for exactly as many colours as the trained data has levels. Qualitative palettes therefore use their first n colours, while sequential and diverging palettes sample n colours across their complete ramps.

Usage

scale_color_biopalette(
  palette,
  ...,
  type = NULL,
  reverse = FALSE,
  palettes_dir = NULL
)

scale_colour_biopalette(
  palette,
  ...,
  type = NULL,
  reverse = FALSE,
  palettes_dir = NULL
)

scale_fill_biopalette(
  palette,
  ...,
  type = NULL,
  reverse = FALSE,
  palettes_dir = NULL
)

Arguments

palette

Character. Name of the palette.

...

Passed to ggplot2::discrete_scale(), for example name, breaks, labels, limits, na.value, drop, and guide.

type

Character. One of "sequential", "diverging", or "qualitative". If NULL, the type is detected automatically.

reverse

Logical. Reverse the palette. Default: FALSE.

palettes_dir

Character. Directory holding a palette collection. If NULL, the palettes bundled with the package are used.

Details

All three palette types are supported. Qualitative palettes suit unordered categories; sequential and diverging palettes can be useful for ordered categories.

Value

A ggplot2 discrete scale.

Examples

library(ggplot2)

ggplot(iris, aes(Sepal.Length, Sepal.Width, colour = Species)) +
  geom_point() +
  scale_color_biopalette("three_body")

ggplot(iris, aes(Species, Sepal.Length, fill = Species)) +
  geom_boxplot() +
  scale_fill_biopalette("three_body", guide = "none")

Continuous ggplot2 Gradients from a biopalette Palette

Description

Build a continuous colour or fill gradient from a sequential or diverging biopalette palette. The complete palette supplies the gradient stops to ggplot2::scale_colour_gradientn() or ggplot2::scale_fill_gradientn(). Gradients are interpolated in Lab colour space, the same space used when get_palette() resamples sequential and diverging palettes.

Usage

scale_color_biopalette_gradient(
  palette,
  ...,
  type = NULL,
  reverse = FALSE,
  values = NULL,
  midpoint = NULL,
  transform = "identity",
  palettes_dir = NULL
)

scale_colour_biopalette_gradient(
  palette,
  ...,
  type = NULL,
  reverse = FALSE,
  values = NULL,
  midpoint = NULL,
  transform = "identity",
  palettes_dir = NULL
)

scale_fill_biopalette_gradient(
  palette,
  ...,
  type = NULL,
  reverse = FALSE,
  values = NULL,
  midpoint = NULL,
  transform = "identity",
  palettes_dir = NULL
)

Arguments

palette

Character. Name of the palette.

...

Passed to ggplot2::scale_colour_gradientn() or ggplot2::scale_fill_gradientn(), for example name, breaks, labels, limits, transform, na.value, and guide.

type

Character. "sequential" or "diverging". If NULL, the type is detected automatically.

reverse

Logical. Reverse the palette. When values is supplied, its spacing is reversed with the colours. Default: FALSE.

values

Optional numeric positions for the palette colours, as accepted by ggplot2::scale_colour_gradientn(). Must have one value per palette colour, be non-decreasing, and lie in ⁠[0, 1]⁠.

midpoint

Optional finite numeric value to place at the visual centre of a diverging palette. Values are rescaled with equal units on either side of this point, so the shorter side of an asymmetric data range does not reach the palette's extreme colour. midpoint cannot be combined with values or a custom rescaler. The default, NULL, applies ordinary range-based scaling.

transform

A transformation specification accepted by ggplot2::continuous_scale(). The same transformation is applied to midpoint before the gradient is rescaled, so the centre remains correct under transformations such as "log10". Default: "identity".

palettes_dir

Character. Directory holding a palette collection. If NULL, the palettes bundled with the package are used.

Details

Qualitative palettes are rejected because interpolating unordered category colours does not produce a meaningful continuous scale. Use scale_color_biopalette() or scale_fill_biopalette() for those palettes.

Value

A ggplot2 continuous scale.

Examples

library(ggplot2)

ggplot(mtcars, aes(wt, mpg, colour = hp)) +
  geom_point(size = 3) +
  scale_color_biopalette_gradient("mitonuclear_blue")

ggplot(mtcars, aes(factor(cyl), factor(gear), fill = mpg - mean(mpg))) +
  geom_tile() +
  scale_fill_biopalette_gradient("walter_white", midpoint = 0)