R中手动排序并重命名geom_bar堆叠条形图

在使用R绘制geom_bar堆叠条形图时，需实现以下需求：

• 聚类分析得到k=1-3的簇分配，但簇数字标签顺序不符合预期：Cluster A对应1，Cluster B对应3，Cluster C对应2
• 手动控制条形图顺序，让1、3、2分别对应Cluster A、B、C，同时X的填充顺序匹配簇内的堆叠逻辑

用户提供的模拟数据及尝试代码如下：

#LIBRARIES

library(viridis)
library(ggplot2)
library(hrbrthemes)

#MOCK DATA

X <- sample(seq(from = 0, to = 5, by = 1), size = 50, replace = TRUE)

Cluster <- sample(seq(from = 1, to = 3, by = 1), size = 50, replace = TRUE)

DAT %>% mutate(Cluster = case_when(X < 2 ~ "1",
                                   X <4 ~ "3",
                                   X <6 ~ "2"))
X <- as.factor(X)
DAT <- data.frame(X, Cluster)

#EXISTING CODE ATTEMPT

ggplot(DAT, aes(x= Cluster, fill = X), color = "X") +
geom_bar(position = position_fill(reverse = TRUE), stat="count", width = 0.80) +
theme_ipsum(axis = TRUE, grid = FALSE, axis_col = "black", base_size = 9) +
theme(plot.title = element_text(hjust = 0.5, size = 9, face = "bold"),
axis.text = element_text(size = 9)) +
scale_fill_viridis_d(option = "G")+
  (scale_x_reverse(breaks = c(1:3), labels = paste("Cluster", LETTERS[1:3])))+
   coord_flip() +
  xlab("")

期望效果：

核心通过设置因子水平顺序控制条形图和填充逻辑，以下是修正后的完整代码：

# 加载依赖包
library(viridis)
library(ggplot2)
library(hrbrthemes)
library(dplyr)

# 生成可复现的模拟数据
set.seed(123)
X <- sample(seq(from = 0, to = 5, by = 1), size = 50, replace = TRUE)
DAT <- data.frame(X) %>% 
  mutate(Cluster = case_when(
    X < 2 ~ "1",
    X < 4 ~ "3",
    X < 6 ~ "2"
  )) %>% 
  # 强制指定Cluster的显示顺序为1→3→2，X按0-5的自然顺序排列
  mutate(
    Cluster = factor(Cluster, levels = c("1", "3", "2")),
    X = factor(X, levels = as.character(0:5))
  )

# 绘制堆叠条形图
ggplot(DAT, aes(x = Cluster, fill = X)) +
  geom_bar(position = position_fill(reverse = FALSE), stat = "count", width = 0.80) +
  # 为Cluster设置对应标签
  scale_x_discrete(labels = c("Cluster A", "Cluster B", "Cluster C")) +
  # 确保填充色顺序匹配X的因子顺序
  scale_fill_viridis_d(option = "G", breaks = levels(DAT$X)) +
  theme_ipsum(axis = TRUE, grid = FALSE, axis_col = "black", base_size = 9) +
  theme(
    plot.title = element_text(hjust = 0.5, size = 9, face = "bold"),
    axis.text = element_text(size = 9)
  ) +
  coord_flip() +
  xlab("")

• Cluster顺序控制：用factor(Cluster, levels = c("1", "3", "2"))直接定义条形图的排列顺序，替换原代码中错误的scale_x_reverse（Cluster是离散因子，非连续变量）
• X填充顺序匹配：将X转换为有序因子，再通过scale_fill_viridis_d的breaks参数绑定填充色顺序，保证堆叠逻辑和类别顺序一致
• 数据修正：修复原代码中DAT未定义就调用mutate的错误，用链式语法简化数据生成流程
• 堆叠方向调整：将position_fill(reverse = TRUE)改为reverse = FALSE，让X的堆叠顺序和因子顺序对齐（可根据实际需求反向）

内容的提问来源于stack exchange，提问作者EB3112