R language - color selection and setting

In the use of R When drawing colors , You have to choose for a long time every time , Sometimes I forget the relevant records , I hereby summarize some commonly used .

1、R base Color version

1.1 R base Color

R The common color code is enough for the color setting in the language ,RGB Format 、16 Base format , Or use relevant color characters .
Refer to the link below to choose the color , But this one only displays characters .

http://www.stat.columbia.edu/~tzheng/files/Rcolor.pdf

In addition, I saw a blog full of colors , Relevant color codes are recorded ：

https://blog.csdn.net/bone_ace/article/details/47362619

When we call multiple colors, we can use rainbow():

plot(rep(1,50),col=rainbow(50), pch=19,cex=2)

1.2 R ggplot Default color inversion

ggplot Default color matching and color flipping （reversed colors）,scale_color_hue and scales In bag hue_pal()(n) ：

library(gridExtra) ##  Typesetting picture 
library(ggplot2) ## ggplot mapping 
library(scales) ##  call hue_pal function 

##  Reference resources 
https://stackoverflow.com/questions/45868625/how-to-reverse-the-default-color-palette-for-ggplot2
##  The next two have the same effect , Default scale_color_hue(direction = 1)
p1 <- ggplot(iris, aes(x=Sepal.Length, y=Sepal.Width, color=Species))+
  geom_point(size=2) + theme_bw()
p2 <- ggplot(iris, aes(x=Sepal.Length, y=Sepal.Width, color=Species)) +
  geom_point(size=2) + theme_bw() + scale_color_hue(direction = 1)
 ## direction = -1  Will make the color reverse . 
##  You need to adjust the starting point to make the reverse color order consistent , Use  h.start=90
p3 <- ggplot(iris, aes(x=Sepal.Length, y=Sepal.Width, color=Species))+
  geom_point(size=2)+  scale_color_hue(direction = -1, h.start=90) + theme_bw()


# Get the colors with 3 classes
cols <- hue_pal()(3)  

# Plot the data and reverse the color
p4 <- ggplot(iris, aes(x=Sepal.Length, y=Sepal.Width, color=Species)) +
  geom_point(size=2) + scale_color_manual(values = rev(cols)) + theme_bw()

grid.arrange(p1, p2,p3,p4, 
             ncol = 2, nrow = 2)

As you can see from the diagram , The color setting of the picture above is opposite to the setting below , Default base R Use the color inversion in rev(cols) that will do .

2、R Color Pack RColorBrewer

Actually , The color above can already meet our color selection needs , But when choosing color matching , We may not match , Cause the picture to look strange .
therefore , It is wiser to use some developed themes ：

library(RColorBrewer)
par(mar=c(3,4,2,2))
display.brewer.all()
br_pal <- brewer.pal(11,"RdYlBu")  ##  Extract the color value of a certain color 
br_pal
[1] "#A50026" "#D73027" "#F46D43" "#FDAE61" "#FEE090" "#FFFFBF" "#E0F3F8" "#ABD9E9" "#74ADD1"
[10] "#4575B4" "#313695"

The three color sets in the following figure correspond to ：sequential、qualitative、diverging

In addition, these colors can be found on interactive websites colorbrewer2 Search for , Website colorbrewer2（https://colorbrewer2.org） It also includes the following three different modes （sequential、qualitative、diverging）, There are different options to query and select color codes .

Besides , The following is also a R Packages can be selected , But this depends on your choice , Enter the following command , A clickable selection box will appear , Not very convenient .

library("colorspace")
pal <- choose_palette()

3、R Color Pack wesanderson、viridis、ggsci

3.1 wesanderson package

## github Address 
https://github.com/karthik/wesanderson
require(wesanderson)
names(wes_palettes)
 [1] "BottleRocket1"  "BottleRocket2"  "Rushmore1"      "Rushmore"       "Royal1"        
 [6] "Royal2"         "Zissou1"        "Darjeeling1"    "Darjeeling2"    "Chevalier1"    
[11] "FantasticFox1"  "Moonrise1"      "Moonrise2"      "Moonrise3"      "Cavalcanti1"   
[16] "GrandBudapest1" "GrandBudapest2" "IsleofDogs1"    "IsleofDogs2"

col = wes_palette("Zissou1", 10, type = "continuous")

height = 1:10
barplot(height, border = "white",col=wes_palette("Zissou1", 10, type = "continuous"),space = 0)

3.2 viridis package

##  Reference link 
https://cran.r-project.org/web/packages/viridis/vignettes/intro-to-viridis.html
require(viridis)

Here is viridis Several color themes of the bag , It can be used as a common color list Input ：

viridis(10) ## viridis  Topic extraction 10 One color 
inferno(10) ## inferno  Topic extraction 10 One color 
plot(1:10, col=viridis(10),cex=4,pch=20)

ggplot You can also use this color theme directly scale_color_viridis and scale_fill_viridis：

p <- ggplot(mtcars, aes(wt, mpg))
p + geom_point(size=4, aes(colour = factor(cyl))) +
    scale_color_viridis(discrete=T,option = "inferno") +
    theme_bw()

3.3 ggsci package

ggsci It should be only ggplot Color matching ：

##  Reference resources ggsci link 
https://cran.r-project.org/web/packages/ggsci/vignettes/ggsci.html

4、R Function to generate multiple continuous colors

Sometimes you need to use multiple colors ,eg,50 When it comes to color ,, Default rainbow Color doesn't like , At this time, you can consider function generation .

4.1 wesanderson and viridis Function to generate multiple consecutive colors

require(wesanderson)
wes_palette("Zissou1", 5, type = "discrete")
wes_palette("Zissou1", 50, type = "continuous")

require(viridis)
viridis(50)
 [1] "#440154FF" "#46085CFF" "#471064FF" "#48176AFF" "#481F70FF" "#482576FF" "#472C7AFF" "#46337EFF"
 [9] "#443983FF" "#423F85FF" "#404588FF" "#3E4A89FF" "#3C508BFF" "#39568CFF" "#365C8DFF" "#34618DFF"
[17] "#31668EFF" "#2F6B8EFF" "#2D718EFF" "#2B758EFF" "#297A8EFF" "#277F8EFF" "#25848EFF" "#23898EFF"
[25] "#218E8DFF" "#20928CFF" "#1F978BFF" "#1E9D89FF" "#1FA187FF" "#21A685FF" "#25AB82FF" "#29AF7FFF"
[33] "#30B57CFF" "#38B977FF" "#40BD72FF" "#4AC16DFF" "#55C568FF" "#5FCA61FF" "#6BCD5AFF" "#77D153FF"
[41] "#84D44BFF" "#91D742FF" "#9FDA3AFF" "#ACDC30FF" "#BADE28FF" "#C8E020FF" "#D6E21AFF" "#E4E419FF"
[49] "#F1E51DFF" "#FDE725FF"
barplot(rep(50,50),col=viridis(50),space=0,border = viridis(50))

4.2 Use colorRampPalette Get multiple consecutive colors

The above function can put some diverged colors Transform to multiple consecutive values , For example, the following RColorBrewer In the function package "Spectral" Only for 11 The most colors , When using continuous pheatmap In the middle of the day , These colors are not enough .
here , have access to colorRampPalette function .

## colorRampPalette The function is system base R The function in , Can be called directly 
colfunc<-colorRampPalette(brewer.pal(11,"Spectral"))
colfunc<-colorRampPalette(c("red","yellow","springgreen","royalblue")) 
plot(rep(1,50),col=(colfunc(50)), pch=19,cex=2)  ##  take 50 One color , The color picture is shown below 
color=colfunc(50) ##  Fetch 50 Continuous color transformation 

The above records only .

1. http://blog.csdn.net/bone_ace/article/details/47362619 （ All colors ）
2. http://www.stat.columbia.edu/~tzheng/files/Rcolor.pdf （ All colors ）
3. https://r-graph-gallery.com/38-rcolorbrewers-palettes.html （RColorBrewer）
4. https://colorbrewer2.org/#type=qualitative&scheme=Set3&n=12（colorbrewer2 Website ）
5. https://www.nceas.ucsb.edu/sites/default/files/2020-04/colorPaletteCheatsheet.pdf
6. https://stackoverflow.com/questions/45868625/how-to-reverse-the-default-color-palette-for-ggplot2 （ Default ggplot Color reversal ）