R 数据可视化 —— ggplot 注释

前言

通常，我们画完图之后可能需要为图形添加注释，如添加参考线、添加文本标签或者添加形状等。

各种注释都有对应的函数来绘制，如我们之前使用过的函数 geom_abline、geom_text、geom_rect 等

或者使用 annotate 函数，并通过参数 geom 来设置对应的注释类型

示例

1. 添加文本

例如，在指定位置添加文本

p <- ggplot(mtcars, aes(x = wt, y = mpg)) + geom_point()
p + annotate("text", x = 4, y = 25, label = "text")

如果想要显示中文，需要设置字体，否则会乱码。如

p <- ggplot(mtcars, aes(x = wt, y = mpg)) + geom_point()
p + annotate("text", x = 4, y = 25, label = "注释", family = 'STKaiti')

常用字体中英文对照

目前测试能用的字体有（MAC 系统）：

• STKaiti
• STFangsong
• STSong
• STXihei

在多个位置添加文本

p + annotate("text", x = 2:5, y = 25, label = "注释", family = "STKaiti")

添加表达式

p + annotate("text", x = 4, y = 25, label = "italic(R) ^ 2 == .75",
             parse = TRUE)
# 或
p + annotate("text", x = 4, y = 25,
             label = "paste(italic(R) ^ 2, \" = .75\")", parse = TRUE)

或者使用 geom_text

p <- ggplot(mtcars, aes(wt, mpg, label = rownames(mtcars)))

p + geom_text()

避免重叠

p + geom_text(check_overlap = TRUE)

或者使用带背景框的 geom_label

p + geom_label()

添加表达式

p + geom_text(aes(label = paste(wt, "^(", cyl, ")", sep = "")),
              parse = TRUE, family = "Times New Roman")

2. 添加线条

添加线段

p + annotate("segment", x = 2.5, xend = 4, y = 15, yend = 25,
             colour = "#66c2a5")

添加误差线

p1 <- p + annotate("pointrange", x = 3.5, y = 20, ymin = 12, ymax = 28,
             colour = "#e41a1c", size = 1)

p2 <- p + annotate("errorbar", x = 3.5, y = 20, ymin = 12, ymax = 28,
             colour = "#377eb8", size = 1)

p3 <- p + annotate("linerange", x = 3.5, y = 20, ymin = 12, ymax = 28,
             colour = "#4daf4a", size = 1)

p4 <- p + annotate("crossbar", x = 3.5, y = 20, ymin = 12, ymax = 28,
             colour = "#984ea3", size = 1)

plot_grid(p1, p2, p3, p4, 
          labels = c("pointrange", "errorbar", "linerange", "crossbar"))

3. 添加形状

绘制矩形

p + annotate("rect", xmin = 3, xmax = 4.2, ymin = 12, ymax = 21,
             fill = "#66c2a5", colour = "black", alpha = .2)

添加椭圆

ggplot(faithful, aes(waiting, eruptions)) +
  geom_point() +
  stat_ellipse()

分组数据椭圆

ggplot(faithful, aes(waiting, eruptions, color = eruptions > 3)) +
  geom_point() +
  stat_ellipse()

设置不同的分布类型，默认是 "t" 假设数据为多变量 t 分布

ggplot(faithful, aes(waiting, eruptions, color = eruptions > 3)) +
  geom_point() +
  stat_ellipse(type = "norm", linetype = 2) +
  stat_ellipse(type = "t") +
  scale_color_manual(values = c("#984ea3", "#377eb8"))

如果想要绘制圆形，需要与 coord_fixed 搭配使用，圆半径为 level

ggplot(faithful, aes(waiting, eruptions, color = eruptions > 3)) +
  geom_point() +
  stat_ellipse(type = "norm", linetype = 2) +
  stat_ellipse(type = "euclid", level = 3) +
  coord_fixed()

stat_ellipse 默认使用的是对象是 path，也可以使用多边形填充

ggplot(faithful, aes(waiting, eruptions, fill = eruptions > 3)) +
  stat_ellipse(geom = "polygon") +
  geom_point(shape = 21)

那如果我想绘制分组数据的边界，要怎么办呢？

我们可以先对数据进行分组，然后使用 chull 函数计算出每组数据的凸点，再将每个分组的凸点合并为数据框

注意：我们应该先对数据进行拆分，因为 chull 函数返回的是点的索引，对数据 group_by 之后，索引是基于每个分组而不是整个数据而言的，切记。

df <- group_by(faithful, eruptions > 3) %>%
  group_split() %>%
  lapply(., function (x) x[chull(x$waiting, x$eruptions),]) %>%
  do.call(rbind, .)

得到了所有凸点之后，就可以绘制数据边界了

ggplot(faithful, aes(waiting, eruptions, fill = eruptions > 3)) +
  geom_polygon(data = df, aes(waiting, eruptions), alpha = 0.5, colour = "grey40") +
  geom_point(shape = 21, alpha = 0.6) +
  scale_fill_manual(values = c("#66c2a5", "#fc8d62")) +
  theme_bw()