Part1boxjitter | 完美复刻Nature上的高颜值统计图

11. 写在前面

原文地址: https://www.nature.com/articles/nature25479

今天要复刻的是这张Nature上的高颜值统计图，一半box，一半jitter

22. 用到的包

rm(list = ls())
library(tidyverse)
library(cowplot)
library(ggparl)

33. 示例数据

dat <- read.csv("geom_boxjitter_example.csv",row.names = 1)

长这个样子 ᑋᵉᑊᑊᵒ ᵕ̈ ᑋᵉᑊᑊᵒ

44. 数据整理

这里我们先把数据整理成ggplot需要的格式

@(｡･o･)@

dat_long <- dat %>%  
  gather(key, value, 1:6) %>% 
  mutate(loc = factor(loc, levels = c("abro", "dome")),
         type = factor(type),
         key = factor(key))

55. 开始画图

5.1 初步绘图

p1 <- ggplot(dat_long, aes(x = type, y = value, fill = key))+
  geom_boxjitter(outlier.color = NA, 
                 jitter.shape = 21, 
                 jitter.color = NA, 
                 jitter.height = 0.05, 
                 jitter.width = 0.075, 
                 errorbar.draw = TRUE) 
p1

5.2 减少分组

原图只有四个组，我们在这里用filter函数过滤一下，变成四组

dat_long <- dat_long %>% 
  filter(key %in% c("p1", "p2","p3","p4"))
  
p2 <- ggplot(dat_long,
             aes(x = type, y = value, fill = key))+
  geom_boxjitter(outlier.color = NA, 
                 jitter.shape = 21, 
                 jitter.color = NA, 
                 jitter.height = 0.05, 
                 jitter.width = 0.075, 
                 errorbar.draw = TRUE) 
p2

5.3 修改细节

去掉背景,更改颜色

p3 <- p2 +
  theme_bw()+
  theme(panel.grid = element_blank(),
        panel.border = element_blank(),
        axis.line = element_line(colour = "black"),
        legend.position = "none") +
  ylim(c(-0.1, 1.05)) + 
  scale_fill_manual(values = c("#0E72BA", "#D95426",
                               "#ecb21e", "#812e91"))

p3

5.4 只展示其中比较好看的两组

我果然是个颜狗啊 /ᐠ｡ꞈ｡ᐟ\

p4 <- ggplot(
  dat_long %>% filter(key %in% c("p1", "p2")), 
  aes(x = type, y = value, fill = key)) +
  geom_boxjitter(outlier.color = NA, jitter.shape = 21, jitter.color = NA, 
                 jitter.height = 0.05, jitter.width = 0.075, errorbar.draw = TRUE) +
  theme_bw()+
  theme(panel.grid = element_blank(),
        panel.border = element_blank(),
        axis.line = element_line(colour = "black"),
        legend.position = "none") +
  ylim(c(-0.05, 1.05)) + 
  scale_fill_manual(values = c("#ecb21e", "#812e91"))
p4

5.5 加上统计值

这里我们用ggpubr包的stat_compare_means()函数给神图加上统计值

library(ggpubr)
# 添加p值
p5 <- p4 + 
  stat_compare_means(#label = "p.signif", 
                     #label.x = 1.5, 
                     method = "t.test",
                     label.y = 1) 

p5

66. 最后把上面的结果都拼到一起吧

plot_grid(p1, p2, p3, p5, 
          labels = "AUTO",
          label_size = 30)

最后祝大家早日不卷!~

需要示例数据的小伙伴，在公众号回复boxjitter获取吧

点个在看吧各位~ ✐.ɴɪᴄᴇ ᴅᴀʏ 〰