用“空”槽格式化ggplot2圖例文本

我正在嘗試在下面的圖中格式化圖例。這是一些虛擬資料和我的代碼：

df <- data.frame(supplement = c("Control", 
                            "0.5 mM NA", "5 mM NA", "10 mM NA",
                            "0.5 mM NI", "5 mM NAI", "10 mM NI"),
             axis1 = c(1, 2, 3, 4, 5, 6, 7),
             axis2 = c(7, 6, 5, 4, 3, 2, 1))

df %>%
  mutate(supplement = factor(supplement,
                             levels = c("Control", 
                                        "0.5 mM NA", "5 mM NA", "10 mM NA",
                                        "0.5 mM NI", "5 mM NAI", "10 mM NI"))) %>%
  ggplot(aes(x = axis1, y = axis2))  
  geom_point(aes(color = supplement), size = 4)  
  labs(y = "Axis 2",
       x = "Axis 1")  
  scale_colour_manual(name = "",
                      breaks = c("Control", 
                                 "0.5 mM NA", "5 mM NA", "10 mM NA",
                                 "0.5 mM NI", "5 mM NAI", "10 mM NI"),
                      labels = c("No Treatment",
                                 "0.5 mM Nitrate", "5 mM Nitrate", "10 mM Nitrate",
                                 "0.5 mM Nitrite", "5 mM Nitrite", "10 mM Nitrite"),
                      values =  c("#003F5C",
                                  "#824ED4", "#31B3F7", "#FFA600",
                                  "#ABF13D", "#DC267F", "#16F5E3")) 
  guides(color = guide_legend(order = 1, nrow = 2, byrow = TRUE))  
  theme(legend.title = element_blank(),
        legend.text = element_text(size = unit(14, "pt")),
        legend.position = "bottom",
        legend.box = "vertical",
        axis.title = element_text(size = unit(16, "pt")),
        axis.text = element_text(size = unit(14, "pt")),
        legend.key.size = unit(25, "pt"),
        legend.key.width = unit(0, 'mm'))

我得到的是這個。

但我真的很希望他們是這樣的。但是我不知道這在 ggplot 中是否可行，因為我還沒有找到任何可以做到這一點的東西。

uj5u.com熱心網友回復：

如果您可以擺脫傳說中的關鍵背景，那么一種選擇是

1. 進入ncols=3_guide_legend
2. 為您的因素添加兩個額外的類別 為圖例添加兩個額外的插槽或鍵
3. 將這些類別的標簽設定為空字串，顏色等于繪圖背景顏色
4. 另外我們必須設定drop=FALSE為scale_color_manual這樣這些未使用的類別不會被丟棄

library(ggplot2)
library(dplyr)

df <- df %>%
  mutate(supplement = factor(supplement,
                             levels = c("Control", "foo", "bar",
                                        "0.5 mM NA", "5 mM NA", "10 mM NA",
                                        "0.5 mM NI", "5 mM NAI", "10 mM NI")))

ggplot(df, aes(x = axis1, y = axis2))  
  geom_point(aes(color = supplement), size = 4)  
  labs(y = "Axis 2", x = "Axis 1")  
  scale_colour_manual(name = "",
                      breaks = c("Control", "foo", "bar",
                                 "0.5 mM NA", "5 mM NA", "10 mM NA",
                                 "0.5 mM NI", "5 mM NAI", "10 mM NI"),
                      labels = c("No Treatment", "", "",
                                 "0.5 mM Nitrate", "5 mM Nitrate", "10 mM Nitrate",
                                 "0.5 mM Nitrite", "5 mM Nitrite", "10 mM Nitrite"),
                      values =  c("#003F5C", "white", "white",
                                  "#824ED4", "#31B3F7", "#FFA600",
                                  "#ABF13D", "#DC267F", "#16F5E3"),
                      drop = FALSE)  
  guides(color = guide_legend(order = 1, ncol = 3, byrow = TRUE))  
  theme(legend.title = element_blank(),
        legend.text = element_text(size = unit(14, "pt")),
        legend.position = "bottom",
        legend.box = "vertical",
        legend.key = element_rect(fill = NA), 
        axis.title = element_text(size = unit(16, "pt")),
        axis.text = element_text(size = unit(14, "pt")),
        legend.key.size = unit(25, "pt"),
        legend.key.width = unit(0, 'mm'))

uj5u.com熱心網友回復：

我不知道有任何（非復雜）方法可以在單獨的行上拉出一個特定的圖例鍵，但在這種情況下，由于圖例中有 7 個專案，您可以將“不處理”分開通過將指南限制為 3 列并重新排序scale_color_manual()以將控制元件放在最后。

在這種情況下，您的繪圖代碼部分將更改為：

# ... original plot code...
scale_colour_manual(name = "",
                      breaks = c( 
                                 "0.5 mM NA", "5 mM NA", "10 mM NA",
                                 "0.5 mM NI", "5 mM NAI", "10 mM NI", "Control"),
                      labels = c(
                                 "0.5 mM Nitrate", "5 mM Nitrate", "10 mM Nitrate",
                                 "0.5 mM Nitrite", "5 mM Nitrite", "10 mM Nitrite", "No Treatment"),
                      values =  c(
                                  "#824ED4", "#31B3F7", "#FFA600",
                                  "#ABF13D", "#DC267F", "#16F5E3", "#003F5C")) 
  guides(color = guide_legend(order = 1, ncol = 3, byrow = TRUE))  

# ... remaining plot code

另一種方法是使用您的原始代碼，并將圖例放在圖的右側或左側。為此，請洗掉guides(...)line 和legend.position=theme 元素，因為默認放置在右側。

標籤：

上一篇：如何選擇列中最常見的值

下一篇：R中的另一列最大值得到另一列值