ggplot2使用facet_wrap分面时如何设置两套独立填充配色方案？

我当前绘制的图表效果如下：

我需要让柱状图的填充色随facet_wrap的分组变化：简单来说，Colour分面的柱状图使用不同饱和度的彩色填充，Grey分面的柱状图使用不同灰度的灰色填充，期望效果如下：

我可以接受图例条目翻倍，也接受图例命名更具象化（例如R.Group.Colour、R.Group.Grey等），恳请大家提供解决方案。

示例数据 #

df <- structure(list(group = c("R.Group", "R.Group", "R.Group", "R.Group", 
"R.Group", "R.Group", "F.Group", "F.Group", "F.Group", "F.Group", 
"F.Group", "F.Group", "G.Group", "G.Group", "G.Group", "G.Group", 
"G.Group", "G.Group"), mode = c("Grey", "Grey", "Grey", "Colour", 
"Colour", "Colour", "Grey", "Grey", "Grey", "Colour", "Colour", 
"Colour", "Grey", "Grey", "Grey", "Colour", "Colour", "Colour"
), stim = structure(c(1L, 2L, 3L, 1L, 2L, 3L, 1L, 2L, 3L, 1L, 
2L, 3L, 1L, 2L, 3L, 1L, 2L, 3L), .Label = c("Words", "Drawings", 
"Photographs"), class = "factor"), emmean = c(0.100851416515973, 
0.108649789029536, 0.0664556962025316, 0.0411392405063291, 0.0574894514767932, 
0.0274261603375527, 0.378405666063894, 0.23372513562387, 0.139240506329114, 
0.185126582278481, 0.209915611814346, 0.0611814345991561, 0.185299879445449, 
0.100663050030139, 0.110759493670886, 0.147151898734177, 0.0743670886075949, 
0.0569620253164557), ymin = c(0.0807866262301694, 0.0848024457881558, 
0.0468323222930633, 0.0269666440471827, 0.0402160043121709, 0.015560956877645, 
0.344334779798359, 0.202260415849806, 0.11272781055522, 0.155246316858577, 
0.178339882849687, 0.0428311209142308, 0.159640657907614, 0.0791229394712789, 
0.0867398048483564, 0.119923200384746, 0.0551207397946581, 0.0390178805506117
), ymax = c(0.120916206801778, 0.132497132270916, 0.0860790701119999, 
0.0553118369654756, 0.0747628986414156, 0.0392913637974605, 0.412476552329429, 
0.265189855397934, 0.165753202103008, 0.215006847698385, 0.241491340779005, 
0.0795317482840815, 0.210959100983284, 0.122203160588998, 0.134779182493416, 
0.174380597083609, 0.0936134374205317, 0.0749061700822997)), class = "data.frame", row.names = c(NA, 
-18L))

当前绘图代码 #

df.plot <-
    ggplot(df, aes(x = stim, y = emmean, fill = group)) +
    scale_fill_manual("legend", values = c("R.Group" = "#202124", "F.Group" = "#808083","G.Group" = "#C0C0C0")) + 
    geom_col(position = position_dodge()) +
    geom_errorbar(aes(ymin = ymin, ymax = ymax), width=0.2, size=0.3, position = position_dodge(0.9)) +
    facet_wrap(~mode, strip.position = "bottom") +
    scale_y_continuous(limits = c(0, 0.5)) +
    theme_classic() +
    theme(axis.title.x = element_blank(),
          axis.text.x = element_text(size = 9),
          axis.text.y = element_text(size = 9),
          strip.placement = "outside",
          strip.background = element_rect(fill = "white", size = 0.5, linetype = "solid", colour = "black"),
          strip.text = element_text(size = 9),
          plot.margin = margin(t = 1, r = 1, b = 0, l = 1, "lines"))

核心思路是构造一个同时包含分组和分面模式的新变量作为填充映射项，就可以为不同分面的同分组设置完全独立的颜色。你可以直接复用以下修改后的代码：

首先给数据集新增组合变量：

# 新增组合变量，把group和mode拼接为填充映射的维度
df$fill_group <- paste(df$group, df$mode, sep = ".")

然后修改绘图代码：

library(ggplot2)

df.plot <-
  ggplot(df, aes(x = stim, y = emmean, fill = fill_group)) +
  # 自定义颜色，前三个为灰度分面的颜色，后三个为彩色分面的颜色
  scale_fill_manual("分组", 
                    values = c("R.Group.Grey" = "#202124", 
                               "F.Group.Grey" = "#808083",
                               "G.Group.Grey" = "#C0C0C0",
                               "R.Group.Colour" = "#D92121", # 红色对应R组彩色
                               "F.Group.Colour" = "#2176D9", # 蓝色对应F组彩色
                               "G.Group.Colour" = "#21D957"), # 绿色对应G组彩色
                    # 按需修改标签命名即可
                    labels = c("R.Group.Grey" = "R组(灰度)",
                               "F.Group.Grey" = "F组(灰度)",
                               "G.Group.Grey" = "G组(灰度)",
                               "R.Group.Colour" = "R组(彩色)",
                               "F.Group.Colour" = "F组(彩色)",
                               "G.Group.Colour" = "G组(彩色)")) + 
  geom_col(position = position_dodge(0.9)) +
  geom_errorbar(aes(ymin = ymin, ymax = ymax), width=0.2, size=0.3, position = position_dodge(0.9)) +
  facet_wrap(~mode, strip.position = "bottom") +
  scale_y_continuous(limits = c(0, 0.5)) +
  theme_classic() +
  theme(axis.title.x = element_blank(),
        axis.text.x = element_text(size = 9),
        axis.text.y = element_text(size = 9),
        strip.placement = "outside",
        strip.background = element_rect(fill = "white", size = 0.5, linetype = "solid", colour = "black"),
        strip.text = element_text(size = 9),
        plot.margin = margin(t = 1, r = 1, b = 0, l = 1, "lines"))

# 输出图片
print(df.plot)

彩色的色值可以根据你的需求自行调整，图例的命名也可以通过修改scale_fill_manual里的labels参数自定义。

内容的提问来源于stack exchange，提问作者adajam