如何在ggplot分面图的X轴文本中斜体显示部分词汇

需求说明 #

需要将X轴文本中的细菌物种名（如Escherichia coli）设置为斜体，菌株名（如Strain 1）保持常规字体，现有数据和初始绘图代码如下：

数据框定义 #

MyDF <- data.frame(Activity=rep(c("Activity 1", "Activity 2", "Activity 3"), each = 6), 
                   Bacteria = rep(c("Escherichia coli Strain 1", "Escherichia coli Strain 2",
                                "Escherichia coli Strain 3", "Escherichia coli Strain 4",
                                "Escherichia coli Strain 5", "Escherichia coli Strain 6"), each=3),
                   Facet=rep(c("Facet 1", "Facet 1","Facet 2","Facet 2","Facet 2","Facet 2"), each=3),
                   Production=rep(1:18))

初始绘图代码 #

MyPlot<- ggplot(MyDF, aes(x=Bacteria, y=Production, fill= Activity)) + 
  geom_bar(stat="identity", 
           position=position_dodge()) +
  facet_grid(Activity ~ Facet, drop=FALSE, scales="free", space = "free_x")+ 
  theme(axis.text.x = element_text(angle = 55, hjust = 1))

MyPlot

实现步骤 #

核心思路是将X轴标签转换为ggplot可识别的表达式（expression），以此实现混合字体样式：

1. 拆分文本并生成表达式向量 #

先拆分物种名和菌株名，再用italic()函数标记斜体部分，最后转换为表达式：

# 拆分每个细菌条目为物种名和菌株编号
split_bact <- strsplit(MyDF$Bacteria, " Strain ")
species <- sapply(split_bact, "[", 1)
strain_num <- sapply(split_bact, "[", 2)

# 创建带格式的表达式文本，再转换为表达式向量
bact_labels <- paste0("italic('", species, "') ~ 'Strain ", strain_num, "'")
bact_expr <- parse(text = bact_labels)

2. 修改绘图代码替换X轴标签 #

在原绘图代码中添加scale_x_discrete()，将X轴标签替换为生成的表达式：

MyPlot <- ggplot(MyDF, aes(x=Bacteria, y=Production, fill= Activity)) + 
  geom_bar(stat="identity", position=position_dodge()) +
  facet_grid(Activity ~ Facet, drop=FALSE, scales="free", space = "free_x")+
  # 替换X轴标签为带格式的表达式
  scale_x_discrete(labels = bact_expr) +
  theme(axis.text.x = element_text(angle = 55, hjust = 1))

MyPlot

简化版（无需单独拆分变量） #

可以直接在scale_x_discrete的labels参数中定义匿名函数，一步完成文本处理：

MyPlot <- ggplot(MyDF, aes(x=Bacteria, y=Production, fill= Activity)) + 
  geom_bar(stat="identity", position=position_dodge()) +
  facet_grid(Activity ~ Facet, drop=FALSE, scales="free", space = "free_x")+
  scale_x_discrete(labels = function(x) {
    split_x <- strsplit(x, " Strain ")
    parse(text = paste0("italic('", sapply(split_x, "[", 1), "') ~ 'Strain ", sapply(split_x, "[", 2), "'"))
  }) +
  theme(axis.text.x = element_text(angle = 55, hjust = 1))

MyPlot

内容的提问来源于stack exchange，提问作者Najoua