Follow Cell Learning | 3. Box plot + scatter + significant difference test
"Practice is the only test of truth."
"Reproduction is the best way to learn R."
DOI: 10.1016/j.cell.2020.05.032
cellThis omics article on the new crown published in 2020 contains a lot of life letter content. Today I will take you to reproduce one of them Supplemental Figure: a boxplot with scatter .
Code and sample data collection in this article: background reply " 20210416 "
Compare common visualization methods for showing differences.
Data Format
Note: The above data is purely fictitious and has no practical significance!
ideas
Boxplot + Significance Test + Scatter
Common ways to compare means:
| method | R implementation function | describe |
|---|---|---|
| T-test | t.test() | Comparing two groups (parametric test) |
| Wilcoxon test | wilcox.test() | Comparing two groups (nonparametric test) |
| ANOVA | aov() or anova() | Comparing multiple groups (parametric test) |
| Kruskal-Wallis | kruskal.test() | Comparing multiple groups (nonparametric test) |
draw
#------
title: "boxplot"
author: "MZBJ"
date: "2020/4/16"
#-----
# 导入所需的包
library(ggplot2)
library(ggsignif)
library(ggpubr)
library(RColorBrewer)
setwd("F:/HJH/mzbj/cell")
# 导入数据
plot_data <- read.csv(file = "ABCD.csv")
#-----------------------
p<- ggplot(data=plot_data)+
geom_boxplot(mapping=aes(x=group,y=Retive_Abundance,colour = group ), #箱线图
alpha = 0.5,
size=1.5,
width = 0.6)+
geom_jitter(mapping=aes(x=group,y=Retive_Abundance,colour = group), #散点
alpha = 0.3,size=3)+
scale_color_manual(limits=c("A","B","C","D"),
values=c("#85B22E","#5F80B4","#E29827","#922927"))+ #颜色
geom_signif(mapping=aes(x=group,y=Retive_Abundance), # 不同组别的显著性
comparisons = list(c("A", "B"), # 哪些组进行比较
c("A", "C"),
c("A", "D"),
c("B", "C"),
c("B", "D"),
c("C", "D")),
map_signif_level=T, # T显示显著性，F显示p value
tip_length=c(0,0,0,0,0,0,0,0,0,0,0,0), # 修改显著性线两端的长短
y_position = c(40,41,42,39,38,40), # 设置显著性线的位置高度
size=1, # 修改线的粗细
textsize = 4, # 修改显著性标记的大小
test = "t.test")+ # 检验的类型
theme_classic( # 主题设置，这个是无线条主题
base_line_size = 1 # 坐标轴的粗细
)+
labs(title="White blood cell(WBC)",x="",y="Retive_Abundance")+ # 添加标题，x轴，y轴内容
theme(plot.title = element_text(size = 15,
colour = "black",
hjust = 0.5),
axis.title.y = element_text(size = 15,
# family = "myFont",
color = "black",
face = "bold",
vjust = 1.9,
hjust = 0.5,
angle = 90),
legend.title = element_text(color="black", # 修改图例的标题
size=15,
face="bold"),
legend.text = element_text(color="black", # 设置图例标签文字
size = 10,
face = "bold"),
axis.text.x = element_text(size = 13, # 修改X轴上字体大小，
color = "black", # 颜色
face = "bold", # face取值：plain普通，bold加粗，italic斜体，bold.italic斜体加粗
vjust = 0.5, # 位置
hjust = 0.5,
angle = 0), #角度
axis.text.y = element_text(size = 13,
color = "black",
face = "bold",
vjust = 0.5,
hjust = 0.5,
angle = 0)
)
p
You're done!
Note: One flaw is that the y-axis is too long. But after I set the y-axis range, the difference markers are also truncated. The solution that can be thought of at present is that the AI simply pulls down the Y axis.
Past content:
Follow CELL to learn to map | 1. Volcano map
Follow the Cell Science | 2. Histogram + Error Bar + Scatter + Significant Difference Test