如何在R语言径向图中实现柱状顶部文本的精准对齐？

我有如下数据：

data_melted <- structure(list(variable = structure(c(1L, 1L, 1L, 1L, 1L, 1L, 
1L, 2L, 2L, 2L, 2L, 2L, 2L, 2L), levels = c("SJLIFE", "CCSS"), class = "factor"), 
    SN_types = structure(c(1L, 2L, 3L, 4L, 5L, 6L, 7L, 1L, 2L, 
    3L, 4L, 5L, 6L, 7L), levels = c("Any SN", "SMN", "NMSC", 
    "Breast cancer", "Thyroid cancer", "Meningioma", "Sarcoma"
    ), class = "factor"), AF_by = c("Combined", "Combined", "Combined", 
    "Combined", "Combined", "Combined", "Combined", "Combined", 
    "Combined", "Combined", "Combined", "Combined", "Combined", 
    "Combined"), value = c(0.536, 0.467, 0.688, 0.704, 0.866, 
    0.423, 0.316, 0.435, 0.323, 0.577, 0.743, 0.664, 0.444, 0.362
    ), new_value = c(54, 47, 69, 70, 87, 42, 32, 44, 32, 58, 
    74, 66, 44, 36)), row.names = c(6L, 12L, 18L, 24L, 30L, 36L, 
42L, 48L, 54L, 60L, 66L, 72L, 78L, 84L), class = "data.frame")

我用以下代码绘制径向图，但柱子上方的文本始终没法正确对齐。我想要文本位于每个柱子顶部，与柱子呈90度夹角，同时保持径向图的角度位置，请问该怎么调整？

原有代码：

angle_increment.AF <- 360 / 14
angle_increment.SN_type <- 360 / 7    
ggplot(data_melted, aes(x = SN_types, y = new_value, fill = variable, label = new_value)) +
      geom_bar(stat = "identity", position = position_dodge(width = 0.9)) +
      geom_vline(xintercept = seq(1.5, 19.5, by = 3), color = "black", linetype = "dashed") + # If you want to include a vertical line
      geom_hline(yintercept = 100) +
      geom_text(aes(angle = 0 + (angle_increment.AF * (as.numeric(as.factor(SN_types)) - 1))),  # Adjust angle of text labels
                position = position_dodge(width = 1), vjust = 0.9, size = 5, color = "black") +  # Adjust label position and size
      scale_fill_manual(values = c(darker_sjlife, darker_ccss), labels = c("SJLIFE", "CCSS")) +  # Specify colors for SJLIFE and CCSS bars
      scale_y_continuous(breaks = seq(0, 100, by = 10)) +  # Custom grid lines from 0 to 1 by 0.25
      coord_polar() +
      theme_minimal() +
      theme(panel.grid.major.y = element_line(color = "gray", linetype = "dotted"),  # Adjust major grid lines
            panel.grid.minor = element_blank(),  # Remove minor grid lines
            axis.line = element_blank(),   # Remove axis lines
            axis.text.x = element_text(hjust = 0.5, angle = angle_increment, color = "black"),  # Center SN type labels within grid
            axis.text = element_text(size = 12, color = "black"),  # Adjust axis text size
            axis.title = element_text(size = 14, color = "black"),  # Adjust axis title size
            legend.title = element_text(size = 14, color = "black"),  # Adjust legend title size
            legend.text = element_text(size = 12, color = "black")) +  # Adjust legend text size
      labs(title = "",
           x = NULL, y = "(%)",
           fill = "Cohort")

想要实现文本垂直于柱子（90度夹角）且准确定位在柱子顶部，需要从文本角度计算和位置调整两方面修改：

关键调整项 #

• 为每个柱子单独计算旋转角度：每个SN_types分组占360/7度，两个柱子分别向左右偏移，保证角度对应柱子的径向方向
• 调整文本的vjust和hjust参数，让文本贴在柱子顶部，同时避免下半圈文本倒置显示异常
• 修正X轴标签的角度，使其对齐分组中心
• 扩展Y轴上限，防止顶部文本被图表边界截断

完整修改后代码 #

# 先定义颜色（避免原代码中未定义变量报错）
darker_sjlife <- "#2c3e50"
darker_ccss <- "#e74c3c"

# 为每个柱子计算专属角度：分组基础角度 + 柱子偏移角度
data_melted$angle <- (as.numeric(data_melted$SN_types) - 1) * (360/7) + 
                      ifelse(data_melted$variable == "SJLIFE", -180/(7*2), 180/(7*2))

ggplot(data_melted, aes(x = SN_types, y = new_value, fill = variable, label = new_value)) +
  geom_bar(stat = "identity", position = position_dodge(width = 0.9)) +
  geom_vline(xintercept = seq(1.5, 7.5, by = 1), color = "black", linetype = "dashed") + 
  geom_hline(yintercept = 100) +
  geom_text(aes(angle = angle),
            position = position_dodge(width = 0.9), 
            vjust = -0.2, 
            hjust = ifelse(angle > 90 & angle < 270, 1, 0),
            size = 5, color = "black") +
  scale_fill_manual(values = c(darker_sjlife, darker_ccss), labels = c("SJLIFE", "CCSS")) +
  scale_y_continuous(breaks = seq(0, 100, by = 10), limits = c(0, 110)) + # 扩展Y轴避免文本被截断
  coord_polar() +
  theme_minimal() +
  theme(panel.grid.major.y = element_line(color = "gray", linetype = "dotted"),
        panel.grid.minor = element_blank(),
        axis.line = element_blank(),
        axis.text.x = element_text(hjust = 0.5, 
                                  angle = seq(0, 360 - 360/7, length.out = 7)), # X轴标签对应分组中心角度
        axis.text = element_text(size = 12, color = "black"),
        axis.title = element_text(size = 14, color = "black"),
        legend.title = element_text(size = 14, color = "black"),
        legend.text = element_text(size = 12, color = "black")) +
  labs(title = "",
       x = NULL, y = "(%)",
       fill = "Cohort")

代码说明 #

• 角度计算：每个分组的基础角度是(as.numeric(SN_types)-1)*360/7，两个柱子分别偏移±180/(7*2)度，确保文本旋转后垂直于柱子的径向方向
• 文本位置：vjust=-0.2把文本放在柱子顶部上方；hjust根据角度切换，下半圈（90-270度）文本右对齐，避免倒置后显示错乱
• Y轴扩展：limits=c(0,110)给顶部文本留出显示空间，防止被截断
• X轴标签：用seq(0, 360 - 360/7, length.out=7)生成每个分组的中心角度，让标签精准对齐分组位置

内容的提问来源于stack exchange，提问作者Achal Neupane