这种显示多种统计结果的云雨图(Raincloud plots)怎么绘制？？

DataCharm

发布于 2022-02-17 15:06:09

5.6K00

代码可运行

运行总次数：0

代码可运行

今天小编再给大家介绍一个在统计学中常用的图表-云雨图(Raincloud plots)。雨云图本质上是一个混合图，可同时将原始数据、数据分布和关键汇总统计数据可视化表现出来，由对分的小提琴图(Violin plot)、箱线图(boxplot) 和作为某种散点的原始数据组成。接下来，小编就介绍几种绘制该类型图表的绘制方法，详细内容如下：

R-ggdist包绘制

R-ggdist包可以很好的绘制云雨图(Raincloud plots)，主要使用其stat_halfeye() 和ggplot2的geom_boxplot() 和 geom_jitter() 图层，具体如下：

plot01 <- ggplot(iris, aes(Species, Sepal.Width)) + 
  ggdist::stat_halfeye(aes(color=Species,fill=Species),adjust = .5, width = .7, .width = 0, justification = -.2, point_colour = NA) + 
  geom_boxplot(aes(color=Species),width = .2, outlier.shape = NA) + 
  geom_jitter(aes(color=Species),width = .05, alpha = .3) +
  ggsci::scale_color_nejm()+
  ggsci::scale_fill_nejm() +
  coord_flip()+
  labs(
    title = "Example of <span style='color:#D20F26'>ggdist::stat_halfeye function</span>",
    subtitle = "processed charts with <span style='color:#1A73E8'>stat_halfeye()</span>",
    caption = "Visualization by <span style='color:#0057FF'>DataCharm</span>") +
  hrbrthemes::theme_ipsum(base_family = "Roboto Condensed") +
  theme(
    plot.title = element_markdown(hjust = 0.5,vjust = .5,color = "black",
                                  size = 20, margin = margin(t = 1, b = 12)),
    plot.subtitle = element_markdown(hjust = 0,vjust = .5,size=15),
    plot.caption = element_markdown(face = 'bold',size = 12),
    legend.position = "none")

Example01 Of ggdist Raincloud plots made

此外，还可以将geom_jitter()替换成ggdist::stat_dots() 绘制另一种样式的云雨图，如下：

plot02 <- ggplot(iris, aes(Species, Sepal.Width)) + 
  ggdist::stat_halfeye(aes(color=Species,fill=Species),adjust = .5, width = .7, .width = 0, justification = -.2, point_colour = NA) + 
  geom_boxplot(aes(color=Species),width = .2, outlier.shape = NA) + 
  ggdist::stat_dots(aes(color=Species,fill=Species),side = "left", dotsize = .5, justification = 1.1, binwidth = .1)+
  ggsci::scale_color_nejm()+
  ggsci::scale_fill_nejm() +
  labs(
    title = "Example of <span style='color:#D20F26'>ggdist::stat_halfeye function</span>",
    subtitle = "processed charts with <span style='color:#1A73E8'>stat_halfeye()+stat_dots()</span>",
    caption = "Visualization by <span style='color:#0057FF'>DataCharm</span>") +
  hrbrthemes::theme_ipsum(base_family = "Roboto Condensed") +
  theme(
    plot.title = element_markdown(hjust = 0.5,vjust = .5,color = "black",
                                  size = 20, margin = margin(t = 1, b = 12)),
    plot.subtitle = element_markdown(hjust = 0,vjust = .5,size=15),
    plot.caption = element_markdown(face = 'bold',size = 12),
    legend.position = "none")

Example02 Of ggdist Raincloud plots made

更多关于 R-ggdist包的介绍，可参考：R-ggdist包介绍[1]

R-gghalves包优化

由于上述绘制的结果不同类型图表较为集中，这里可以使用R-gghalves包对其布局进行优化，如下：

plot03 <- ggplot(iris, aes(Species, Sepal.Width)) + 
  ggdist::stat_halfeye(aes(color=Species,fill=Species),adjust = .5, width = .7, .width = 0, justification = -.3, point_colour = NA) + 
  geom_boxplot(aes(color=Species),width = .1, outlier.shape = NA) +
  gghalves::geom_half_point(aes(color=Species),side = "l", range_scale = .4, alpha = .5) +
  ggsci::scale_color_nejm()+
  ggsci::scale_fill_nejm() +
  coord_flip()+
  labs(
    title = "Example of <span style='color:#D20F26'>gghalves::geom_half_point function</span>",
    subtitle = "processed charts with <span style='color:#1A73E8'>geom_half_point()</span>",
    caption = "Visualization by <span style='color:#0057FF'>DataCharm</span>") +
  hrbrthemes::theme_ipsum(base_family = "Roboto Condensed") +
  theme(
    plot.title = element_markdown(hjust = 0.5,vjust = .5,color = "black",
                                  size = 20, margin = margin(t = 1, b = 12)),
    plot.subtitle = element_markdown(hjust = 0,vjust = .5,size=15),
    plot.caption = element_markdown(face = 'bold',size = 12),
    legend.position = "none")

Example01 Of ggdist+ gghalves Raincloud plots made

当然，你也可以将更改点的形状，如下：

plot04 <- ggplot(iris, aes(Species, Sepal.Width)) + 
  ggdist::stat_halfeye(aes(color=Species,fill=Species),adjust = .5, width = .5, .width = 0, justification = -.3, point_colour = NA) + 
  geom_boxplot(aes(color=Species),width = .1, outlier.shape = NA) +
  gghalves::geom_half_point(aes(color=Species),side = "l", range_scale = 0, shape = 95, size = 15, alpha = .3) +
  ggsci::scale_color_nejm()+
  ggsci::scale_fill_nejm() +
  labs(
    title = "Example of <span style='color:#D20F26'>gghalves::geom_half_point function</span>",
    subtitle = "processed charts with <span style='color:#1A73E8'>geom_half_point()</span>",
    caption = "Visualization by <span style='color:#0057FF'>DataCharm</span>") +
  hrbrthemes::theme_ipsum(base_family = "Roboto Condensed") +
  theme(
    plot.title = element_markdown(hjust = 0.5,vjust = .5,color = "black",
                                  size = 20, margin = margin(t = 1, b = 12)),
    plot.subtitle = element_markdown(hjust = 0,vjust = .5,size=15),
    plot.caption = element_markdown(face = 'bold',size = 12),
    legend.position = "none")

Example02 Of ggdist+ gghalves Raincloud plots made

更多关于gghalves包的介绍可参考：gghalves包介绍[2]

R-raincloudplots包绘制

这里小编再介绍一个专门绘制云雨图(Raincloud plots) 的第三方包-R-raincloudplots 。该包和上述上述介绍的方法对比：

该包将ggplot功能包装到一个函数中，并不是直接添加geom_或stat_ 绘图层，但其他图层及主题样式还是可以添加的。
该包目前还没发布在CRAN上，需要从Github上进行安装，可能导致一些包的不兼容。

下面就通过几个具体的例子进行展示：

「样例一」：

library(raincloudplots)
df_1x1 <- data_1x1(
  array_1 = iris$Sepal.Length[1:50],
  array_2 = iris$Sepal.Length[51:100],
  jit_distance = .09,
  jit_seed = 321)
mycolor  <-pal_jco()(2)
raincloud_1_h <- raincloud_1x1(
  data = df_1x1, 
  colors = mycolor, 
  fills = mycolor, 
  size = 1, 
  alpha = .8, 
  ort = 'h') +

scale_x_continuous(breaks=c(1,2), labels=c("Group1", "Group2"), limits=c(0, 3)) +
  xlab("Groups") + 
  ylab("Score") +
  labs(
    title = "Example of <span style='color:#D20F26'>raincloudplots::raincloud_1x1 function</span>",
    subtitle = "processed charts with <span style='color:#1A73E8'>raincloud_1x1()</span>",
    caption = "Visualization by <span style='color:#0057FF'>DataCharm</span>") +
  hrbrthemes::theme_ipsum(base_family = "Roboto Condensed") +
  theme(
    plot.title = element_markdown(hjust = 0.5,vjust = .5,color = "black",
                                  size = 20, margin = margin(t = 1, b = 12)),
    plot.subtitle = element_markdown(hjust = 0,vjust = .5,size=15),
    plot.caption = element_markdown(face = 'bold',size = 12),
    legend.position = "none")

Example01 Of raincloudplots::raincloud_1x1()

「样例二」：

raincloud_2 <- raincloud_1x1_repmes(
  data = df_1x1,
  colors = mycolor,
  fills = mycolor,
  line_color = 'gray',
  line_alpha = .6,
  size = 1,
  alpha = .8,
  align_clouds = FALSE) +
 
scale_x_continuous(breaks=c(1,2), labels=c("Pre", "Post"), limits=c(0, 3)) +
  xlab("Time") + 
  ylab("Score") +
  labs(
    title = "Example of <span style='color:#D20F26'>raincloudplots::raincloud_1x1 function</span>",
    subtitle = "processed charts with <span style='color:#1A73E8'>raincloud_1x1()</span>",
    caption = "Visualization by <span style='color:#0057FF'>DataCharm</span>") +
  hrbrthemes::theme_ipsum(base_family = "Roboto Condensed") +
  theme(
    plot.title = element_markdown(hjust = 0.5,vjust = .5,color = "black",
                                  size = 20, margin = margin(t = 1, b = 12)),
    plot.subtitle = element_markdown(hjust = 0,vjust = .5,size=15),
    plot.caption = element_markdown(face = 'bold',size = 12),
    legend.position = "none")