跟着Nature学作图：R语言ggplot2箱线图、小提琴图、抖动散点图

论文是

Environmental factors shaping the gut microbiome in a Dutch population

数据和代码的github主页链接

https://github.com/GRONINGEN-MICROBIOME-CENTRE/DMP

这个也是数据代码的下载链接，可以看目录结构

https://zenodo.org/record/5910709#.YmAcp4VBzic

今天的推文重复一下论文中的figure2c

image.png

论文中提供的代码自定义和一个很长的 函数，好像是把统计检验和作图全都综合到一起了，但是我没看明白其中统计检验的部分，这里就把作图代码单独整理出来了，统计检验的FDR值最后手动添加

首先是读取数据

dfToPlot<-read.csv("dfToPlot.csv")
head(dfToPlot)

给x轴的变量赋予因子水平

dfToPlot$RELATIONSHIP.0 <- factor(dfToPlot$RELATIONSHIP.0,
                                  levels=c("RND.PAIR","PARTNERS","PARENT_CHILD","SIBLINGS"))

这个因子水平主要是控制x轴的先后顺序

加载ggplot2

library(ggplot2)

箱线图

ggplot(data=dfToPlot,aes(x=RELATIONSHIP.0,
                         y=BC_Spec,
                         color=RELATIONSHIP.0))+
  geom_boxplot()

抖动散点图

ggplot(data=dfToPlot,aes(x=RELATIONSHIP.0,
                         y=BC_Spec,
                         color=RELATIONSHIP.0))+
  geom_jitter()

小提琴图

ggplot(data=dfToPlot,aes(x=RELATIONSHIP.0,
                         y=BC_Spec,
                         color=RELATIONSHIP.0))+
  geom_violin()

三个图拼图

library(ggplot2)
ggplot(data=dfToPlot,aes(x=RELATIONSHIP.0,
                         y=BC_Spec,
                         color=RELATIONSHIP.0))+
  geom_boxplot() -> p1

p1

ggplot(data=dfToPlot,aes(x=RELATIONSHIP.0,
                         y=BC_Spec,
                         color=RELATIONSHIP.0))+
  geom_jitter() -> p2
p2

ggplot(data=dfToPlot,aes(x=RELATIONSHIP.0,
                         y=BC_Spec,
                         color=RELATIONSHIP.0))+
  geom_violin() -> p3
p3

image.png

将三个图叠加到一张图上

cbPalette <- c("#E69F00", "#CC79A7", "#56B4E9", "#009E73", "#CC79A7", "#F0E442", "#999999","#0072B2","#D55E00")
ggplot(data=dfToPlot,aes(x=RELATIONSHIP.0,
                         y=BC_Spec,
                         color=RELATIONSHIP.0))+
  geom_jitter(alpha=0.2,
              position=position_jitterdodge(jitter.width = 0.35, 
                                            jitter.height = 0, 
                                            dodge.width = 0.8))+
  geom_boxplot(alpha=0.2,width=0.45,
               position=position_dodge(width=0.8),
               size=0.75,outlier.colour = NA)+
  geom_violin(alpha=0.2,width=0.9,
              position=position_dodge(width=0.8),
              size=0.75)+
  scale_color_manual(values = cbPalette)+
  theme_classic() +
  theme(legend.position="none") + 
  theme(text = element_text(size=16)) + 
  #ylim(0.0,1.3)+
  ylab("Bray-Curtis distance of Species")

image.png

最后是手动添加统计检验的文字

ggplot(data=dfToPlot,aes(x=RELATIONSHIP.0,
                         y=BC_Spec,
                         color=RELATIONSHIP.0))+
  geom_jitter(alpha=0.2,
              position=position_jitterdodge(jitter.width = 0.35, 
                                            jitter.height = 0, 
                                            dodge.width = 0.8))+
  geom_boxplot(alpha=0.2,width=0.45,
               position=position_dodge(width=0.8),
               size=0.75,outlier.colour = NA)+
  geom_violin(alpha=0.2,width=0.9,
              position=position_dodge(width=0.8),
              size=0.75)+
  scale_color_manual(values = cbPalette)+
  theme_classic() +
  theme(legend.position="none") + 
  theme(text = element_text(size=16)) + 
  #ylim(0.0,1.3)+
  ylab("Bray-Curtis distance of Species")+
  #scale_x_discrete(labels=c("A","B","C","D"))+
  annotate("segment", x = 1-0.01, y = 1, xend = 2.01,lineend = "round", 
           yend = 1,size=1,colour="black",arrow = arrow(length = unit(0.02, "npc")))+
  annotate("segment", x = 2.01, y = 1, xend = 0.99,lineend = "round", 
           yend = 1,size=1,colour="black",arrow = arrow(length = unit(0.02, "npc")))+
  annotate("text", x=1.5,y=1.01, 
           label=expression("**"~"FDR"~2.41%*%10^-10),vjust=0)

image.png

他这里的双箭头的处理方式是把一个单箭头添加两次

制作封面图

p4+p4+scale_color_manual(values = cbPalette[5:8])

image.png

今天推文的示例数据和代码可以在公众号后台留言20220505获取

欢迎大家关注我的公众号

小明的数据分析笔记本

小明的数据分析笔记本 公众号 主要分享：1、R语言和python做数据分析和数据可视化的简单小例子；2、园艺植物相关转录组学、基因组学、群体遗传学文献阅读笔记；3、生物信息学入门学习资料及自己的学习笔记！