R ggpubr：添加平均水平线&；分组配对比较？

我最近发现了很棒的

ggpubr

包。当我与多个组一起工作时，我喜欢将数据拆分为facet（

facet.by

）

我的问题由两个子问题组成，与同一个图表相关。我想

添加每组的唯一平均水平线，以及

仅按组绘制显著性比较

我的目标是显示箱线图及其配对比较显著性值。由于各组的显著性可能不同，我可以只绘制显著性组，或者指定显著性比较，并按组分别绘制

示例数据：

library(ggpubr)
library(ggplot2)  

# Create data
# :::::::::::::::::::::::::::::::::::::::::::::::::::
data("ToothGrowth")
df <- ToothGrowth

# Create basic plot
p <- ggboxplot(df, 
               x = "dose", 
               y = "len",
               color = "dose", 
               palette =c("#00AFBB", "#E7B800", "#FC4E07"),
               add = "jitter",
               facet.by = "supp", # define faceting 
               shape = "dose")

# Add horizontal line per each group???
p + geom_hline(yintercept = mean(df$len), #aggregate(len ~ supp, df, mean)$len, # mean(df$len), 
               linetype = 2,
               group = "supp")

显著性结果取决于组。然而，我只想展示重要的比较。如何在每组中仅指定有效对？比如：

my_comp_OJ <- list( c("0.5", "1"), c("0.5", "2"))
my_comp_VC <- list(  c("0.5", "1"), c("1", "2"), c("0.5", "2"))

---

您必须计算每个组的总平均值，并有一个data.frame，表示平均值来自哪个组

> aggregate(len ~ supp, data = df, FUN = mean)
  supp      len
1   OJ 20.66333
2   VC 16.96333

在

geom\u hline

中包含这一点很简单

grand.means <- aggregate(len ~ supp, data = df, FUN = mean)
ggboxplot(df, 
          x = "dose", 
          y = "len",
          color = "dose", 
          palette = c("#00AFBB", "#E7B800", "#FC4E07"),
          add = "jitter",
          facet.by = "supp", # define faceting 
          shape = "dose") +
  geom_hline(data = grand.means, aes(yintercept = len),
             linetype = 2,
             group = "supp")

---

grand.means真棒，谢谢！！我不知道我可以如此轻松地在geom_hline中定义数据源，但这绝对是有意义的！
> aggregate(len ~ supp, data = df, FUN = mean)
  supp      len
1   OJ 20.66333
2   VC 16.96333

grand.means <- aggregate(len ~ supp, data = df, FUN = mean)
ggboxplot(df, 
          x = "dose", 
          y = "len",
          color = "dose", 
          palette = c("#00AFBB", "#E7B800", "#FC4E07"),
          add = "jitter",
          facet.by = "supp", # define faceting 
          shape = "dose") +
  geom_hline(data = grand.means, aes(yintercept = len),
             linetype = 2,
             group = "supp")