Scientific Journal and Sci-Fi Themed Color Palettes for ggplot2

Nan Xiao <http://nanx.me>
Miaozhu Li <http://miaozhu.li>

2016-11-20

1 Introduction

My eyes were finally opened and I understood nature.

I learned at the same time to love it.

— Claude Monet

ggsci offers a collection of high-quality color palettes inspired by colors used in scientific journals, data visualization libraries, and science fiction TV shows. The color palettes in ggsci are available as ggplot2 scales. For all the color palettes, the corresponding scales are named as:

We also provided aliases, such as scale_colour_palname() for scale_color_palname(). All available color palettes are summarized in the table below.

Name Scales Palette Types Palette Generator

NPG

scale_color_npg() scale_fill_npg()

"nrc"

pal_npg()

AAAS

scale_color_aaas() scale_fill_aaas()

"default"

pal_aaas()

Lancet

scale_color_lancet() scale_fill_lancet()

"lanonc"

pal_lancet()

JCO

scale_color_jco() scale_fill_jco()

"default"

pal_jco()

UCSCGB

scale_color_ucscgb() scale_fill_ucscgb()

"default"

pal_ucscgb()

D3

scale_color_d3()
scale_fill_d3()

"category10" "category20" "category20b" "category20c"

pal_d3()

UChicago

scale_color_uchicago() scale_fill_uchicago()

"default"
"light"
"dark"

pal_uchicago()

The Simpsons

scale_color_simpsons() scale_fill_simpsons()

"springfield"

pal_simpsons()

Futurama

scale_color_futurama() scale_fill_futurama()

"planetexpress"

pal_futurama()

Rick and Morty

scale_color_rickandmorty() scale_fill_rickandmorty()

"schwifty"

pal_rickandmorty()

GSEA

scale_color_gsea() scale_fill_gsea()

"default"

pal_gsea()

2 Discrete Color Palettes

We will use scatterplots with smooth curves, and bar plots to demonstrate the discrete color palettes in ggsci.

library("ggsci")
library("ggplot2")
library("gridExtra")

data("diamonds")

p1 = ggplot(subset(diamonds, carat >= 2.2),
       aes(x = table, y = price, colour = cut)) +
  geom_point(alpha = 0.7) +
  geom_smooth(method = "loess", alpha = 0.05, size = 1, span = 1) +
  theme_bw()

p2 = ggplot(subset(diamonds, carat > 2.2 & depth > 55 & depth < 70), 
       aes(x = depth, fill = cut)) +
  geom_histogram(colour = "black", binwidth = 1, position = "dodge") +
  theme_bw()

2.1 NPG

The NPG palette is inspired by the plots in journals published by Nature Publishing Group:

p1_npg = p1 + scale_color_npg()
p2_npg = p2 + scale_fill_npg()
grid.arrange(p1_npg, p2_npg, ncol = 2)

2.2 AAAS

The AAAS palette is inspired by the plots in journals published by American Association for the Advancement of Science:

p1_aaas = p1 + scale_color_aaas()
p2_aaas = p2 + scale_fill_aaas()
grid.arrange(p1_aaas, p2_aaas, ncol = 2)

2.3 Lancet

The Lancet palette is inspired by the plots in Lancet journals, such as Lancet Oncology:

p1_lancet = p1 + scale_color_lancet()
p2_lancet = p2 + scale_fill_lancet()
grid.arrange(p1_lancet, p2_lancet, ncol = 2)

2.4 JCO

The JCO palette is inspired by the the plots in Journal of Clinical Oncology:

p1_jco = p1 + scale_color_jco()
p2_jco = p2 + scale_fill_jco()
grid.arrange(p1_jco, p2_jco, ncol = 2)

2.5 UCSCGB

The UCSCGB palette is from the colors used by UCSC Genome Browser for representing chromosomes. This palette has been intensively used in visualizations produced by Circos.

p1_ucscgb = p1 + scale_color_ucscgb()
p2_ucscgb = p2 + scale_fill_ucscgb()
grid.arrange(p1_ucscgb, p2_ucscgb, ncol = 2)

2.6 D3

The D3 palette is from the categorical colors used by D3.js (version 3.x and before). There are four palette types (category10, category20, category20b, category20c) available.

p1_d3 = p1 + scale_color_d3()
p2_d3 = p2 + scale_fill_d3()
p1_d3_c10 = p1 + scale_color_d3("category10")
p2_d3_c10 = p2 + scale_fill_d3("category10")
p1_d3_c20 = p1 + scale_color_d3("category20")
p2_d3_c20 = p2 + scale_fill_d3("category20")
p1_d3_c20b = p1 + scale_color_d3("category20b")
p2_d3_c20b = p2 + scale_fill_d3("category20b")
p1_d3_c20c = p1 + scale_color_d3("category20c")
p2_d3_c20c = p2 + scale_fill_d3("category20c")
grid.arrange(p1_d3_c10,  p2_d3_c10,
             p1_d3_c20,  p2_d3_c20,
             p1_d3_c20b, p2_d3_c20b,
             p1_d3_c20c, p2_d3_c20c,
             ncol = 2, nrow = 4)

2.7 UChicago

The UChicago palette is based on the colors used by The University of Chicago. There are three palette types (default, light, dark) available.

p1_uchicago = p1 + scale_color_uchicago()
p2_uchicago = p2 + scale_fill_uchicago()
p1_uchicago_light = p1 + scale_color_uchicago("light")
p2_uchicago_light = p2 + scale_fill_uchicago("light")
p1_uchicago_dark  = p1 + scale_color_uchicago("dark")
p2_uchicago_dark  = p2 + scale_fill_uchicago("dark")
grid.arrange(p1_uchicago,       p2_uchicago,
             p1_uchicago_light, p2_uchicago_light,
             p1_uchicago_dark,  p2_uchicago_dark,
             ncol = 2, nrow = 3)

2.8 The Simpsons

This palette is inspired by the colors used in the TV show The Simpsons:

p1_simpsons = p1 + scale_color_simpsons()
p2_simpsons = p2 + scale_fill_simpsons()
grid.arrange(p1_simpsons, p2_simpsons, ncol = 2)

2.9 Futurama

This palette is inspired by the colors used in the TV show Futurama:

p1_futurama = p1 + scale_color_futurama()
p2_futurama = p2 + scale_fill_futurama()
grid.arrange(p1_futurama, p2_futurama, ncol = 2)

2.10 Rick and Morty

This palette is inspired by the colors used in the TV show Rick and Morty:

p1_rickandmorty = p1 + scale_color_rickandmorty()
p2_rickandmorty = p2 + scale_fill_rickandmorty()
grid.arrange(p1_rickandmorty, p2_rickandmorty, ncol = 2)

3 Continuous Color Palettes

We will use a correlation matrix visualization (a special type of heatmap) to demonstrate the continuous color palettes in ggsci.

library("reshape2")

data("mtcars")
cor = cor(mtcars)
cor_melt = melt(cor)

p3 = ggplot(cor_melt,
            aes(x = Var1, y = Var2, fill = value)) + 
  geom_tile(colour = "black", size = 0.3) +
  theme_bw()

3.1 GSEA

The GSEA palette (continuous) is inspired by the heatmaps generated by GSEA GenePattern.

p3_gsea = p3 + scale_fill_gsea()
p3_gsea

4 Non-ggplot2 Graphics

To apply the color palettes in ggsci to other graphics systems (such as base graphics and lattice graphics), simply use the palette generator functions in the table above. For example:

mypal = pal_npg("nrc", alpha = 0.7)(9)
mypal
## [1] "#E64B35B2" "#4DBBD5B2" "#00A087B2" "#3C5488B2" "#F39B7FB2" "#8491B4B2"
## [7] "#91D1C2B2" "#DC0000B2" "#7E6148B2"
library("scales")
show_col(mypal)

You will be able to use the generated hex color codes for such graphics systems accordingly. The transparent level of the entire palette is easily adjustable via the argument "alpha" in every generator or scale function.

5 Discussion

Please note some of the palettes might not be the best choice for certain purposes, such as color-blind safe, photocopy safe, or print friendly. If you do have such considerations, you might want to check out color palettes like ColorBrewer and viridis.

The color palettes in this package are solely created for research purposes. The authors are not responsible for the usage of such palettes.

Project website: ggsci.net