如何在R中绘制带纹理的多步骤黑白冲积图？

数据准备 #

首先定义原始数据框：

df <- data.frame(Variable = c("X1", "X2", "X3", "X4", "X5", "X6"),
                 Pearson1 = c(6, 3, 2, 5, 4, 1),
                 Spearman1 = c(6, 5, 1, 2, 3, 4),
                 Kendall1 = c(6, 5, 1, 2, 3, 4),
                 Pearson2 = c(6, 5, 1, 2, 3, 4),
                 Spearman2 = c(6, 5, 1, 2, 4, 3),
                 Kendall2 = c(6, 5, 1, 2, 3, 4))

原代码问题分析 #

你之前的代码无法实现需求，核心问题有三点：

• 错误使用freq = "value"：此处value是排名值，并非冲积图所需的频数统计，应该基于变量的分组关系构建连接
• 未指定步骤顺序，无法将Variable列为第一个展示步骤
• 未实现黑白风格下用纹理/线型区分变量的需求

可行解决方案 #

推荐使用ggalluvial（基于ggplot2，自定义性更强）来实现需求，以下是完整步骤：

1. 数据格式转换 #

将宽格式数据转为适合冲积图的长格式，并明确步骤顺序：

library(tidyverse)
library(ggalluvial)

df_long <- df %>%
  # 转换为长格式，拆分类别与排名
  pivot_longer(cols = -Variable, names_to = "Category", values_to = "Rank") %>%
  # 添加步骤序号，确保Variable为第一步
  mutate(Step = case_when(
    Category == "Pearson1" ~ 2,
    Category == "Spearman1" ~ 3,
    Category == "Kendall1" ~ 4,
    Category == "Pearson2" ~ 5,
    Category == "Spearman2" ~ 6,
    Category == "Kendall2" ~ 7
  )) %>%
  # 补充Variable作为第一步的行数据
  bind_rows(df %>% mutate(Category = "Variable", Rank = Variable, Step = 1)) %>%
  # 按步骤和变量排序，确保连接关系正确
  arrange(Step, Variable)

2. 黑白风格冲积图实现 #

提供两种黑白区分变量的方案：

方案1：用线型区分变量（基础黑白风格）

ggplot(df_long,
       aes(x = Step, stratum = Rank, alluvium = Variable)) +
  # 绘制冲积带，用不同线型区分变量，黑白填充
  geom_alluvium(aes(linetype = Variable), fill = "white", colour = "black") +
  # 绘制分层节点，保持黑白风格
  geom_stratum(fill = "white", colour = "black") +
  # 添加节点标签
  geom_text(stat = "stratum", aes(label = Rank)) +
  # 设置步骤名称
  scale_x_continuous(breaks = 1:7,
                     labels = c("Variable", "Pearson1", "Spearman1", "Kendall1",
                                "Pearson2", "Spearman2", "Kendall2")) +
  # 黑白主题，去掉多余网格
  theme_bw() +
  theme(legend.position = "bottom",
        panel.grid = element_blank()) +
  # 为6个变量设置不同线型
  scale_linetype_manual(values = c("solid", "dashed", "dotted", "dotdash", "longdash", "twodash")) +
  labs(x = "步骤", y = "")

方案2：用纹理区分变量（更直观的黑白差异）

如果需要更明显的区分效果，可使用ggpattern包添加纹理：

library(ggpattern)

ggplot(df_long,
       aes(x = Step, stratum = Rank, alluvium = Variable)) +
  # 带纹理的冲积带，用不同纹理区分变量
  geom_alluvium_pattern(
    aes(pattern = Variable),
    colour = "black",
    fill = "white",
    pattern_fill = "black",
    pattern_density = 0.1,
    pattern_spacing = 0.02
  ) +
  geom_stratum(fill = "white", colour = "black") +
  geom_text(stat = "stratum", aes(label = Rank)) +
  scale_x_continuous(breaks = 1:7,
                     labels = c("Variable", "Pearson1", "Spearman1", "Kendall1",
                                "Pearson2", "Spearman2", "Kendall2")) +
  theme_bw() +
  theme(legend.position = "bottom",
        panel.grid = element_blank()) +
  # 为6个变量设置不同纹理
  scale_pattern_manual(values = c("stripe", "crosshatch", "dot", "circle", "hex", "wave")) +
  labs(x = "步骤", y = "")

内容的提问来源于stack exchange，提问作者nickolakis