r - 使用 ggplot2 在左侧和右侧创建一个带有两个 x 轴的中央 y 轴图

Question

我正在尝试复制我在一篇论文中发现的精彩情节。该图基本上有一个中心 y 轴，然后根据变量在左侧和右侧创建条形图。接下来是剧情:

我有一个类似的数据框来复制这个图。我的数据框 df具有下一个结构:

df
# A tibble: 64 x 4
   Gender Cases Deaths Group
   <chr>  <int>  <dbl> <chr>
 1 F       6163     24 G1   
 2 M       9067    136 G1   
 3 F        430      4 G2   
 4 M       1026     51 G2   
 5 F         43      0 G3   
 6 M         67      1 G3   
 7 F       1382     43 G4   
 8 M        888     26 G4   
 9 F        249      4 G5   
10 M        191     10 G5   
# ... with 54 more rows

我想创建上面提到的图，在中心轴上显示变量 Group在 x 轴(左和右)中，我想显示变量 Cases和 Deaths根据 Gender变量，这将是 M 左边的条形图F 的性别和权利障碍性别。

为了达到目标情节，我草拟了一些可能作为基础的代码，但我不知道如何修改它以更改轴的顺序。这是代码:

library(ggplot2)
library(tidyverse)
#Code for F gender
df %>% pivot_longer(-c(Group,Gender)) %>%
  filter(Gender=='F') %>%
  mutate(Group=factor(Group,levels = unique(df$Group),ordered = T)) %>%
  ggplot(aes(x=Group,y=value,fill=name))+
  geom_bar(stat = 'identity')+
  scale_x_discrete(limits = rev(unique(df$Group)))+
  coord_flip()+
  theme(legend.position = 'top')
#Code for F gender
df %>% pivot_longer(-c(Group,Gender)) %>%
  filter(Gender=='M') %>%
  mutate(Group=factor(Group,levels = unique(df$Group),ordered = T)) %>%
  ggplot(aes(x=Group,y=value,fill=name))+
  geom_bar(stat = 'identity')+
  scale_x_discrete(limits = rev(unique(df$Group)))+
  coord_flip()+
  theme(legend.position = 'top')

产生下一个图:

我在这些图中发现的一个问题是，尽管按照我的意愿显示了条形图，但它们有两个问题。首先，我希望其中一个条形图(例如死亡)可以比另一个条形图(例如案例)更细。其次，柱的结构正在累积所有值。我希望这两个度量在每个轴上都从零开始。

我希望这个情节可以在ggplot2中创作.我的数据df接下来是:

#Data
df <- structure(list(Gender = c("F", "M", "F", "M", "F", "M", "F", 
"M", "F", "M", "F", "M", "F", "M", "F", "M", "F", "M", "F", "M", 
"F", "M", "F", "M", "F", "M", "F", "M", "F", "M", "F", "M", "F", 
"M", "F", "M", "F", "M", "F", "M", "F", "M", "F", "M", "F", "M", 
"F", "M", "F", "M", "F", "M", "F", "M", "F", "M", "F", "M", "F", 
"M", "F", "M", "F", "M"), Cases = c(6163L, 9067L, 430L, 1026L, 
43L, 67L, 1382L, 888L, 249L, 191L, 278L, 248L, 36L, 2925L, 4248L, 
2286L, 1576L, 12106L, 441L, 690L, 7L, 53L, 7645L, 5335L, 10862L, 
2546L, 229L, 9136L, 1578L, 2657L, 9301L, 11384L, 27773L, 34435L, 
23599L, 32952L, 1105L, 170L, 31L, 94L, 3469L, 2408L, 1L, 6L, 
86L, 566L, 1108L, 8355L, 3203L, 7174L, 2314L, 2943L, 46L, 54L, 
26L, 187L, 44201L, 837L, 40608L, 38616L, 343L, 12284L, 6571L, 
8882L), Deaths = c(24, 136, 4, 51, 0, 1, 43, 26, 4, 10, 0, 2, 
1, 242, 84, 112, 49, 1164, 7, 33, 0, 4, 26, 115, 63, 24, 7, 556, 
14, 86, 228, 784, 596, 1344, 189, 705, 24, 15, 0, 1, 180, 120, 
0, 0, 0, 7, 8, 155, 465, 1630, 39, 125, 3, 3, 0, 0, 2511, 87, 
114, 219, 8, 847, 340, 760), Group = c("G1", "G1", "G2", "G2", 
"G3", "G3", "G4", "G4", "G5", "G5", "G6", "G6", "G7", "G7", "G8", 
"G8", "G9", "G9", "G10", "G10", "G11", "G11", "G12", "G12", "G13", 
"G13", "G14", "G14", "G15", "G15", "G16", "G16", "G17", "G17", 
"G18", "G18", "G19", "G19", "G20", "G20", "G21", "G21", "G22", 
"G22", "G23", "G23", "G24", "G24", "G25", "G25", "G26", "G26", 
"G27", "G27", "G28", "G28", "G29", "G29", "G30", "G30", "G31", 
"G31", "G32", "G32")), row.names = c(NA, -64L), class = c("tbl_df", 
"tbl", "data.frame"))

非常感谢!

Answer 1

我根本不会在这里转。如果您将病例和死亡分开，则对 geom_col 的两次调用将重叠。要获得具有中心 y 轴的镜像设置，我会简单地绘制一个矩形和标签作为注释。这意味着您还需要伪造 x 轴，以调整矩形占用的空间。

您需要将其中一种性别的 x 值设置为负值。由于我们需要对数转换，因此在获取对数后将它们设为负值。

df %>%
  mutate(Cases = ifelse(Gender == "F", -log10(Cases) - 0.5, log10(Cases) + 0.5),
         Deaths = ifelse(Deaths == 0, 1, Deaths),
         Deaths = ifelse(Gender == "F", -log10(Deaths) - 0.5, 
                         log10(Deaths) + 0.5)) %>%
  ggplot(aes(Cases, Group, fill = Gender)) +
  geom_col(width = 0.8, alpha = 0.2) +
  geom_col(width = 0.4, aes(x = Deaths)) +
  annotate(geom = "rect", xmin = -0.5, xmax = 0.5, ymax = Inf, ymin = -Inf,
           fill = "white") +
  annotate(geom = "text", x = 0, y = levels(factor(df$Group)),
           label = levels(factor(df$Group))) +
  theme_light() +
  scale_x_continuous(breaks = c(-4.5, -3.5, -2.5, -1.5, 1.5, 2.5, 3.5, 4.5),
                     labels = c("10,000", "1,000", "100", "10", "10",
                                "100", "1,000", "10,000")) +
  scale_fill_manual(values = c("#fc7477", "#6f71f2")) +
  labs(title = "Female                              Male",
       x = "Cases (Deaths shown in solid bars)") +
  theme(axis.text.y = element_blank(),
        axis.title.y = element_blank(),
        axis.ticks.y = element_blank(),
        panel.grid.major.y = element_blank(),
        plot.title = element_text(size = 16, hjust = 0.5),
        legend.position = "none")