如何在geom_point散点图中为特定文本列值添加条件geom_rect高亮

问题背景 #

现有如下数据框和散点图代码：

df <-
    setNames(data.frame(
        as.POSIXct(
            c(
            "2022-07-29 00:00:00",
            "2022-07-29 00:00:05",
            "2022-07-29 00:05:00",
            "2022-07-29 00:05:05",
            "2022-07-29 00:10:00",
            "2022-07-29 00:15:00",
            "2022-07-29 00:20:00",
            "2022-07-29 00:20:05"
            )),
        c(1, 2, 3, 4, 5, 6, 7, 8),
        c(0.8, 2.1, 2.5, 4.1, 5, 6.1, 6.9, 8.1),
        c("a", "a", "b", "b", "b", "b", "b", "c")
    ),
    c("timeStamp", "value1", "value2", "text"))

df %>% ggplot(aes(timeStamp, value1, color =text)) + geom_point()

需要高亮text值为"b"的对应时间区域，尝试的代码无法运行，且必须使用矩形高亮（不能仅修改点的颜色/大小）。

解决方案 #

核心思路是先从数据中提取目标text对应的时间区间，再用geom_rect绘制覆盖该区间的高亮矩形，注意避免继承主绘图的映射关系，防止错误。

完整可运行代码 #

library(tidyverse)

df <-
    setNames(data.frame(
        as.POSIXct(
            c(
            "2022-07-29 00:00:00",
            "2022-07-29 00:00:05",
            "2022-07-29 00:05:00",
            "2022-07-29 00:05:05",
            "2022-07-29 00:10:00",
            "2022-07-29 00:15:00",
            "2022-07-29 00:20:00",
            "2022-07-29 00:20:05"
            )),
        c(1, 2, 3, 4, 5, 6, 7, 8),
        c(0.8, 2.1, 2.5, 4.1, 5, 6.1, 6.9, 8.1),
        c("a", "a", "b", "b", "b", "b", "b", "c")
    ),
    c("timeStamp", "value1", "value2", "text"))

# 提取text="b"对应的时间区间，设置y轴覆盖整个范围
b_time_range <- df %>% 
  filter(text == "b") %>% 
  summarise(
    xmin = min(timeStamp),
    xmax = max(timeStamp),
    ymin = -Inf,
    ymax = Inf
  )

# 绘制图形：先画高亮矩形，再画数据点（避免矩形遮挡点）
df %>% 
  ggplot(aes(timeStamp, value1, color = text)) +
  geom_rect(
    data = b_time_range,
    aes(xmin = xmin, xmax = xmax, ymin = ymin, ymax = ymax),
    fill = "yellow",
    alpha = 0.2,
    inherit.aes = FALSE  # 禁用继承主绘图的映射，防止color映射冲突
  ) +
  geom_point()

代码说明 #

• 提取时间区间：通过filter和summarise精准获取text="b"的最小/最大时间戳，确保高亮范围准确匹配数据。
• 覆盖整个y轴：用-Inf和Inf作为y轴的上下限，无需手动计算数据的极值，保证矩形覆盖整个绘图区域的垂直范围。
• 避免映射冲突：设置inherit.aes = FALSE，让geom_rect不继承主绘图的color=text映射，防止出现错误。
• 图层顺序：先绘制高亮矩形，再绘制数据点，确保点不会被矩形遮挡。

如果需要高亮多个text值，只需扩展时间区间的计算，比如：

# 高亮text为"a"和"b"的区域
multi_time_range <- df %>% 
  filter(text %in% c("a", "b")) %>% 
  group_by(text) %>% 
  summarise(
    xmin = min(timeStamp),
    xmax = max(timeStamp),
    ymin = -Inf,
    ymax = Inf,
    fill_color = case_when(text == "a" ~ "lightblue", text == "b" ~ "yellow")
  )

df %>% 
  ggplot(aes(timeStamp, value1, color = text)) +
  geom_rect(
    data = multi_time_range,
    aes(xmin = xmin, xmax = xmax, ymin = ymin, ymax = ymax, fill = fill_color),
    alpha = 0.2,
    inherit.aes = FALSE,
    show.legend = FALSE
  ) +
  geom_point()

内容的提问来源于stack exchange，提问作者ambergris