如何在ggplot2中根据列值条件设置scale_shape_manual参数？

核心问题 #

scale_shape_manual()的values参数需要接收命名向量（或与因子水平顺序匹配的向量）来定义映射规则，而case_when()是逐行处理数据的函数，无法直接生成符合要求的映射关系；另外代码中未定义unicode对象，也会导致报错。

修正后的代码 #

library(ggplot2)
library(dplyr)

adrs4 <- structure(list(subjid = structure(c(1L, 1L, 1L, 1L, 1L, 1L), levels = c("1", "2", "3", "4", "5", "6", "7", "8", "9", "10", "11", "12", "13", "14", "15", "16", "17", "18", "19", "20", "21", "22", "23", "24", "25", "26", "27", "28", "29", "30", "31"), class = "factor"), 
                        ady = c(38, 84, 111, 111, 242, 242), avalc = structure(c(3L, 4L, 6L, 6L, 7L, 7L), levels = c("CR", "PR", "SD", "PD", "On-treatment", "EOT", "Death"), class = "factor"), 
                        trtedy = c(0, 0, 0, 0, 0, 242), trtpn = c(1L, 1L, 1L, 1L, 1L, 1L), trtp = c("TRT A", "TRT A", "TRT A", "TRT A", "TRT A", "TRT A")), 
                   row.names = c(NA, -6L), class = c("tbl_df", "tbl", "data.frame"))

adrs3_all <- adrs4

ggplot((adrs4 %>% filter(trtpn==1)),aes(reorder(subjid,trtedy), trtedy)) + 
  geom_bar(stat="identity", aes(fill=factor(trtp))) +
  geom_point(data=(adrs3_all %>% filter(trtpn==1)), 
             aes(subjid, ady, colour=avalc, shape=avalc), size=10) +
  coord_flip() +
  theme(text=element_text(family="sans")) +
  scale_color_manual(values = c('darkcyan','brown','darkviolet','red','forestgreen','darkgoldenrod2','black')) +
  # 用命名向量定义avalc各值对应的形状编码
  scale_shape_manual(values = c(
    "CR" = 17,        # 实心上三角
    "PR" = 16,        # 实心圆
    "SD" = 1,         # 空心圆
    "PD" = 22,        # 空心正方形
    "On-treatment" = 25, # 空心上三角
    "EOT" = 8,        # 八角星
    "Death" = 4       # 叉号
  ), guide = "none") +
  scale_y_continuous(limits = c(0,600),breaks=seq(0,600,50)) +
  labs(fill="Treatment", colour="Response", shape="Response", 
       x="Subject ID ", y="Days",
       title="Swimmers Plot",subtitle = 'All Treated Subjects') +
  theme(plot.title = element_text(hjust = 0.5), # 修正居中参数，原1.5会导致标题偏移
        plot.caption = element_text(size=15, hjust=0.5),
        legend.text = element_text(size=15)) +
  theme_classic(base_size = 20)

关键说明 #

1. 形状映射规则：通过命名向量直接指定每个avalc因子水平对应的形状编码，ggplot会自动完成匹配；你可以根据需求替换为其他形状编码（参考ggplot官方形状编码表）
2. 替换unicode对象：使用ggplot内置的形状编码（数字形式）替代未定义的unicode对象，避免报错
3. 修正标题居中：将plot.title的hjust参数从1.5改为0.5，实现标题真正居中

内容的提问来源于stack exchange，提问作者jkatam