如何在ggplot2的geom_point中突出指定分组标记

需要在散点图中突出显示特定国家组：将标记为"U.S. Market"的非洲对美重要市场国家前置显示，同时淡化其他国家。目前已通过geom_point的alpha参数实现其他国家的淡化，但无法让"U.S. Market"组的点前置。

尝试过用ifelse函数处理geom_point（像处理geom_text那样），也试过数据子集化，但出现美学长度错误。以下是当前代码和数据：

当前代码 #

ggplot(HC,aes(W.RGDP,WHCI,label=HC$`Country Name`))+geom_point(aes(fill=`Comparison 2`),size=8,pch=21,alpha=0.4)+
  labs(x="Real GDP Per Capita (log)",y="Human Capital Index")+
  ggtitle("Human Capital Index & GDP Per Capita", subtitle="SSA & World (by income group)")+
  stat_smooth(method="glm",se=TRUE,color="firebrick2")+geom_text(check_overlap=T,family="serif")+
  geom_text(aes(label=ifelse(HC$`Comparison 2`=="U.S. Market",paste0(HC$`Country Name`),"")),family="serif")+
  theme_stata(base_size = 14,base_family="serif")+scale_x_continuous(labels=dollar_format())+
  guides(color=guide_legend(override.aes=list(size=7)))+
  theme(legend.title=element_blank(),legend.position=c(0.12,0.75),
        legend.background=element_rect(colour=NA,fill=NA),
        panel.border=element_rect(color="black",fill="NA"),
        plot.subtitle=element_text(face="italic",size=14))+
  scale_fill_manual(values=c("#1F78B4","#A6CEE3","#33A02C","#B2DF8A","darksalmon","gold3"),
                    breaks=c("High income","Upper middle income","Lower middle income","Low income","Africa","U.S. Market"))

数据集 #

structure(list(`Country Name` = c("Afghanistan", "Albania", "Algeria", 
"Angola", "Antigua and Barbuda", "Argentina", "Armenia", "Australia", 
"Austria", "Azerbaijan", "Bahrain", "Bangladesh", "Belarus", 
"Belgium", "Benin", "Bhutan", "Bosnia and Herzegovina", "Botswana", 
"Brazil", "Brunei Darussalam"), `Comparison 2` = c("Low income", 
"Upper middle income", "U.S. Market", "U.S. Market", "High income", 
"Upper middle income", "Upper middle income", "High income", 
"High income", "Upper middle income", "High income", "Lower middle income", 
"Upper middle income", "High income", "Africa", "Lower middle income", 
"Upper middle income", "Africa", "Upper middle income", "High income"
), `HCI 2020` = c(0.400284141302109, 0.634250938892365, 0.534556448459625, 
0.362405389547348, 0.595703899860382, 0.602144718170166, 0.578998804092407, 
0.770239531993866, 0.746739745140076, 0.577767550945282, 0.652472019195557, 
0.46403643488884, 0.700078189373016, 0.76041966676712, 0.400290727615356, 
0.475297033786774, 0.57968407869339, 0.414028495550156, 0.551497936248779, 
0.626391351222992), `Real GDP Per Capita (3-year average) (log)` = c(2.63994355604787, 
3.68217708824842, 3.65453362326108, 3.38066789845535, 4.20466171351866, 
4.08791987813632, 3.66586025270983, 4.77398337179505, 4.65329694576918, 
3.72765661725707, 4.36960984791812, 3.22739691265819, 3.79971743950937, 
4.63045716032335, 3.08331073406556, 3.50531148290221, 3.77030705526165, 
3.79820120774174, 3.93248670657043, 4.47241060052674)), row.names = c(NA, 
-20L), class = c("tbl_df", "tbl", "data.frame"))

核心思路是分层绘制散点：先画所有非"U.S. Market"的点（设置低透明度淡化），再单独画"U.S. Market"的点（设置全透明度），利用ggplot的图层顺序（后绘制的图层会覆盖前置）实现需求，同时修正数据引用方式避免美学长度错误。

关键修改点 #

• 拆分geom_point为两个独立图层，严格控制绘制顺序
• 用filter函数对数据子集化，确保每个图层只处理对应分组的数据
• 避免在aes中直接用HC$引用列，改用列名直接引用，防止长度不匹配
• 简化冗余的文本图层逻辑，强化视觉突出效果

修正后的完整代码 #

# 先对齐数据列名与代码中的引用
library(dplyr)
HC <- HC %>% 
  rename(W.RGDP = `Real GDP Per Capita (3-year average) (log)`,
         WHCI = `HCI 2020`)

ggplot(HC, aes(W.RGDP, WHCI, label = `Country Name`)) +
  # 1. 绘制非U.S. Market的点，淡化显示
  geom_point(data = filter(HC, `Comparison 2` != "U.S. Market"),
             aes(fill = `Comparison 2`), size = 8, pch = 21, alpha = 0.4) +
  # 2. 绘制U.S. Market的点，前置显示，全透明度
  geom_point(data = filter(HC, `Comparison 2` == "U.S. Market"),
             aes(fill = `Comparison 2`), size = 8, pch = 21, alpha = 1) +
  labs(x = "Real GDP Per Capita (log)", y = "Human Capital Index") +
  ggtitle("Human Capital Index & GDP Per Capita", subtitle = "SSA & World (by income group)") +
  stat_smooth(method = "glm", se = TRUE, color = "firebrick2") +
  # 仅显示U.S. Market的国家名称，强化突出
  geom_text(data = filter(HC, `Comparison 2` == "U.S. Market"),
            family = "serif") +
  # 可选：其他国家文本淡化显示，不需要可删除该行
  geom_text(data = filter(HC, `Comparison 2` != "U.S. Market"),
            check_overlap = TRUE, family = "serif", alpha = 0.7) +
  theme_stata(base_size = 14, base_family = "serif") +
  scale_x_continuous(labels = dollar_format()) +
  guides(fill = guide_legend(override.aes = list(size = 7))) +
  theme(legend.title = element_blank(),
        legend.position = c(0.12, 0.75),
        legend.background = element_rect(colour = NA, fill = NA),
        panel.border = element_rect(color = "black", fill = NA),
        plot.subtitle = element_text(face = "italic", size = 14)) +
  scale_fill_manual(values = c("#1F78B4", "#A6CEE3", "#33A02C", "#B2DF8A", "darksalmon", "gold3"),
                    breaks = c("High income", "Upper middle income", "Lower middle income", "Low income", "Africa", "U.S. Market"))

效果说明 #

• 后绘制的"U.S. Market"点会自动覆盖在其他点上方，实现前置显示
• 分层处理数据避免了美学长度不匹配问题
• 文本图层也可针对性设置，进一步强化核心组的视觉焦点

内容的提问来源于stack exchange，提问作者AdamG66