ggplot2作分组箱线图并添加均值点连线及显著性程度标注
2018-03-19 本文已影响167人
生信杂谈
下面这个箱线图由两部分组成，左边是1,2两组A,B,C,D,E属性的箱线图，右边是1,2两组整体水平的箱线图，下面通过R语言的ggplot2包从头绘制类似这种箱线图及其各个组的均值点连线和显著性程度标注。
载入所小的包并生成数据：
rm(list=ls())
library(data.table)
library(ggplot2)
library(ggsignif)
Group1<-data.frame(A=runif(100,15,90),B=rnorm(100,30,3),C=runif(100,30,60),D=rexp(100,0.1),E=rpois(100,10),group=1)
Group2<-data.frame(A=runif(100,0,100),B=rnorm(100,50,2),C=runif(100,40,70),D=rexp(100,0.3),E=rpois(100,20),group=2)
# 整合数据到一个数据框：
b<-rbind(Group1,Group2)
```R
数据如下，一共有A,B,C,D,E5个属性，分1,2两个组。
将数据框宽格式变长格式方便制图：
```R
b<-melt(b,id.vars = c("group"))
b$group<-as.factor(b$group)
变成长格式的数据：
由于要做每个箱线图的均值点及均值连线，需要获得每个组每个属性的均值，并且定义每个组每个属性的X坐标为固定值。
# group1的mean:
c<-copy(b)
setDF(c)
c1<-tapply(c[c$group==1,"value"],c[c$group==1,"variable"],mean)
c2<-tapply(c[c$group==2,"value"],c[c$group==2,"variable"],mean)
c3<-rbind(data.frame(variable=names(c1),value=c1,group=1),data.frame(variable=names(c2),value=c2,group=2))
c3$group<-as.factor(c3$group)
# 分别计算两组均值，用来画折线图：
c3$variable2<-NA
c3[c3$group==1&c3$variable=="A","variable2"]<-0.795
c3[c3$group==1&c3$variable=="B","variable2"]<-1.795
c3[c3$group==1&c3$variable=="C","variable2"]<-2.795
c3[c3$group==1&c3$variable=="D","variable2"]<-3.795
c3[c3$group==1&c3$variable=="E","variable2"]<-4.795
c3[c3$group==2&c3$variable=="A","variable2"]<-1.185
c3[c3$group==2&c3$variable=="B","variable2"]<-2.185
c3[c3$group==2&c3$variable=="C","variable2"]<-3.185
c3[c3$group==2&c3$variable=="D","variable2"]<-4.185
c3[c3$group==2&c3$variable=="E","variable2"]<-5.185
用来做均值点和连线的数据如下：
开始做两组每个属性的箱线图：
p1<-ggplot(b)+
geom_boxplot(aes(x=variable,y=value,fill=group),width=0.6,position = position_dodge(0.8),outlier.size = 0,outlier.color = "white")+
scale_fill_manual(values = c("red", "blue"),breaks=c("1","2"),labels=c("Group 1","Group 2"))+
geom_point(data=c3,aes(x=variable2,y=value,color=group),shape=15,size=1)+
geom_line(data=c3,aes(x=variable2,y=value,color=group),size=1,linetype = "dotted")+
# geom_smooth(data=c3,aes(x=variable2,y=value,color=group),size=1,linetype = "dashed")+
# stat_summary(fun.y = mean, geom = "errorbar", aes(x=variable,y=value,ymax = ..y.., ymin = ..y..,color=group),width = .75, linetype = "dashed")
xlab("")+
ylab("")+
scale_y_continuous(limits = c(0,110),breaks=seq(0,110,5)) +
geom_signif(stat="identity",
data=data.frame(x=c(0.795,1.795,2.795,3.795,4.795), xend=c(1.185, 2.185,3.185,4.185,5.185),
y=c(106,66,70,30,35), annotation=c("***", " *** "," *** "," ** ","*")),
aes(x=x,xend=xend, y=y, yend=y, annotation=annotation)) +
theme_bw()+
theme(
legend.position = "top",
legend.background=element_blank(),
legend.key = element_blank(),
legend.margin=margin(0,0,0,0,"mm"),
axis.text.x=element_text(size=rel(1.1),face="bold"),
axis.line.x = element_line(size = 0.5, colour = "black"),
axis.line.y = element_line(size = 0.5, colour = "black"),
legend.text=element_text(size=rel(1.1)),
legend.title=element_blank(),
panel.border = element_blank(),
panel.grid = element_blank()
)+
guides(color=FALSE)
p1
生成图像如下：
然后做两组整体水平的箱线图：
p2<-ggplot(b)+
geom_boxplot(aes(x=group,y=value,fill=group),width=0.8,position=position_dodge(1))+
stat_summary(fun.y = mean, geom = "point", aes(x=group,y=value,color=group),shape=15)+
scale_fill_manual(values = c("red", "blue"),breaks=c("1","2"),labels=c("Group 1","Group 2"))+
scale_x_discrete(breaks=c("1","2"),labels=c("Group 1","Group 2"))+
scale_y_continuous(limits = c(0,110),breaks=seq(0,110,5)) +
geom_signif(stat="identity",
data=data.frame(x=c(1), xend=c(2),
y=c(106), annotation=c("**")),
aes(x=x,xend=xend, y=y, yend=y, annotation=annotation))+
theme_bw()+
theme(
legend.position = "top",
legend.background=element_blank(),
legend.key = element_blank(),
legend.margin=margin(0,0,0,0,"mm"),
axis.text.x=element_text(size=rel(1.1),face="bold"),
axis.line.x = element_line(size = 0.5, colour = "black"),
axis.line.y = element_blank(),
axis.ticks.y = element_blank(),
axis.text.y = element_blank(),
axis.title = element_blank(),
legend.text=element_text(size=rel(1.1)),
legend.title=element_blank(),
plot.margin = margin(11.5,0,7,0,"mm"),
panel.border = element_blank(),
panel.grid = element_blank()
)+
guides(fill=F,color=F)
p2
最后合并两个图像：
# 如果没有安装则install.packages("gridExtra")
library(gridExtra)
# 合并两个图：
grid.arrange(p1, p2, nrow=1, ncol=2,widths=c(3.5,1),heights=c(4))
更多原创精彩视频敬请关注生信杂谈:
