ggplot2/ggpubr包：绘制配对箱线图、配对小提琴图和配对点图

数据和代码获取：请查看主页个人信息！！！

关键词“配对图”

大家好，今天我将介绍如何使用R语言ggplot2包和ggpubr包绘制分组/分面配对图的方法。

配对图最大的优点在于可以展示相同样本处理前后对应关系：

图片来源：Nat Commun：Clinical practices underlie COVID-19 patient respiratory microbiome composition and its interactions with the host

Step1：载入数据​​​​​​​

rm(list=ls())pacman::p_load(tidyverse,reshape2,ggpubr)# 实战data <- read.csv('data.csv')head(data)

数据信息：ID是TCGA的barcode，Group是肿瘤与对照样本，paired是配对的barcode信息，Expression则是目标基因的表达量。

Step2：ggpubr包简单示例​​​​​​​

# ggpubr绘图示例ggpaired(ToothGrowth,          x = "supp",          y = "len",         color = "supp",          line.color = "gray",          line.size = 0.4,         palette = "jco") +  stat_compare_means(paired = TRUE)

绘图实战：

Step3：ggpubr包绘制配对箱线图​​​​​​​

ggpaired(data, x = "Group", y = "Expression", color = "black", fill = "Group", line.size = 0.5) +  geom_line(aes(group = paired), color = "grey80") +  stat_compare_means(paired = TRUE, size = 2) +  facet_wrap(~ Type, scales = 'free_x', nrow = 2) +  ggsci::scale_fill_npg() +  ggsci::scale_color_npg() +  theme_classic2(base_size = 9) +  xlab('Expression')ggsave('pic1.png', width = 10, height = 4)

Step4：ggplot2绘制配对箱线图​​​​​​​

facet_theme <- theme(strip.background = element_blank(),                     panel.spacing.x  = unit(0, "cm"), #x轴分面距离                     panel.spacing.y  = unit(0, "cm"),#y轴分面距离                     strip.text.x = element_text(size=7),                     strip.text.y = element_text(size=7, face="bold"),                     axis.text.x=element_text(angle=45,vjust=1, hjust=1),                     legend.position = 'none')
ggplot(data, aes(x = Group, y = Expression, color = Group)) +  geom_boxplot(outlier.size = 0.5) +  geom_line(aes(group = paired), color = "grey80", size = 0.5) +  geom_point(size = 0.5) +  stat_compare_means(comparisons = list(c("tumor", "normal")),                      paired = T, size = 2,label = "p.format",tip.length = 0) +  facet_wrap(~ Type, scales = 'free_y', nrow = 2) +  ggsci::scale_color_npg() +  theme_classic2(base_size = 9) +  facet_theme +  xlab('Expression')ggsave('pic2.png', width = 10, height = 4)

这里优化了分面绘图主题，此外需要注意添加图层的顺序：

geom_boxplot → geom_line → geom_point；

geom_boxplot(outlier.size = 0.5) 和 geom_point(size = 0.5)

这两个函数中散点大小需要保持一致。

Step5：配对小提琴图​​​​​​​

ggplot(data, aes(x = Group, y = Expression, color = Group)) +  geom_violin(size = 1) +  geom_line(aes(group = paired), color = "grey80", size = 0.5) +  geom_point(size = 0.5, color = 'black') +  stat_compare_means(comparisons = list(c("tumor", "normal")),                      paired = T, size = 2,label = "p.format",tip.length = 0) +  facet_wrap(~ Type, scales = 'free_y', nrow = 2) +  ggsci::scale_color_npg() +  theme_classic2(base_size = 9) +  facet_theme +  xlab('Expression')ggsave('pic3.png', width = 10, height = 4)

这里的分组差异展示方式为p值：

Step6：配对点图

ggplot(data, aes(x = Group, y = Expression, fill = Group, color = Group)) +  geom_line(aes(group = paired), color = "grey80", size = 0.5) +  geom_point(size = 1) +  stat_compare_means(comparisons = list(c("tumor", "normal")),                      paired = T, size = 2,label = "p.signif",tip.length = 0) +  facet_wrap(~ Type, scales = 'free_y', nrow = 2) +  scale_color_manual(values = c('red', 'black')) +  theme_classic2(base_size = 12) +  facet_themeggsave('pic4.png', width = 10, height = 4)

这里的分组差异展示方式为*号

• ns: p > 0.05  *: p <= 0.05  **: p <= 0.01  ***: p <= 0.001  ****: p <= 0.0001

数据和代码获取：请查看主页个人信息！！！

关键词“配对图”