在R语言美国地图上绘制带分类的散点饼图（求助分类颜色显示）

你当前代码用的是geom_point而非scatterpie包的专属函数，因此无法实现饼图分类的颜色区分。以下是修正方案：

关键调整点 #

• 加载scatterpie包
• 将散点饼数据转换为宽格式（每个区域一行，各分类作为单独列）
• 使用geom_scatterpie绘制饼图，通过fill映射分类并自定义颜色

完整修正代码 #

library(sf)
library(ggplot2)
library(maps)
library(tidyverse)
library(scatterpie) # 新增加载scatterpie包

# 加载美国地图数据
us_map <- st_as_sf(maps::map("state", plot = FALSE, fill = TRUE))

regions_data <- list(
  Northeast = c("DE", "ME", "MD", "MA", "NJ", "CT", "NY", "PA"),
  OhioValley =  c("IL", "IN", "KY", "MO", "OH", "TN", "WV"),
  RockiesPlains = c("MT", "NE", "ND", "SD", "WY"),
  South = c("AR", "KS", "LA", "MS", "OK", "TX"),
  Northwest = c("WA", "OR", "ID"),
  Southwest = c("AZ", "CO", "NM", "UT"),
  West = "CA",
  Southeast = c("AL", "GA", "FL", "NC", "SC", "VA"),
  UpperMidwest =  c("IA", "MN", "WI", "MI")
)

mapping_df <- enframe(regions_data, name = "region", value = "state") %>%
  unnest(state) %>%
  left_join(data.frame(state.abb, state.name), by = c("state" = "state.abb")) %>%
  mutate(state.name = tolower(state.name))

us_map_regions <- left_join(us_map, mapping_df, by = c("ID" = "state.name"))

# 整理散点饼数据：转换为宽格式
scatterpie_data <- data.frame(
  region = c("Northeast", "Northeast", "Northeast", "OhioValley", "OhioValley", "OhioValley"),
  category = rep(c("Diploid", "Polyploid", "Unknown"), each = 2),
  value = c(30, 40, 30, 20, 30, 50),
  lon = c(-75, -75, -75, -90, -90, -90),
  lat = c(42, 42, 42, 38, 38, 38)
) %>%
  pivot_wider(names_from = category, values_from = value) %>%
  distinct(region, lon, lat, .keep_all = TRUE) # 去重保留每个区域的唯一坐标和数值

# 绘制地图+散点饼图
ggplot() +
  geom_sf(data = us_map_regions, aes(fill = region), alpha = 0.7) +
  theme_minimal() +
  labs(title = "美国区域地图与散点饼图", subtitle = "自定义区域划分") +
  theme(legend.position = "bottom") +
  geom_sf_text(data = us_map_regions, aes(label = region), color = "black", size = 3) +
  # 使用scatterpie绘制饼图
  geom_scatterpie(data = scatterpie_data, 
                  aes(x = lon, y = lat, r = 1.5), # r控制饼图半径
                  cols = c("Diploid", "Polyploid", "Unknown"), # 指定饼图的分类列
                  alpha = 0.8) +
  # 自定义分类颜色
  scale_fill_manual(name = "倍性分类",
                    values = c(Diploid = "#1f77b4", Polyploid = "#ff7f0e", Unknown = "#2ca02c")) +
  coord_sf(datum = NA)

代码解释 #

1. 数据格式转换：用pivot_wider将长格式数据转为宽格式，确保每个区域对应一行，方便scatterpie识别各分类的数值
2. geom_scatterpie参数：cols指定参与饼图绘制的分类列，r控制饼图大小；通过fill映射分类（自动关联scale_fill_manual的颜色）
3. 颜色自定义：scale_fill_manual为每个分类设置独立颜色，同时修改图例名称

内容的提问来源于stack exchange，提问作者PloidyPhD