ggpubr包绘制图表-最全教程

转自  nikang3148  CSDN

提示：

• 阅读本文需要有一定的R语言和ggplot2基础，不熟悉ggplot2语法，可能会存在困难

• 部分代码段用到了管道操作,最好是熟悉tidyverse包，以及阅读过《R for Data Science》

  文章目录

• 执行ggplot函数

• 图例输出

• 点线形式输出

• 调色板输出get palette

• 均值比较

• 分组描述统计desc_statby()

• 增加网格线add gridlines

• 增加分面

• 增加文字标签

• 在已有图形上增加辅助统计量se,sd等

• 添加置信区间换图

• 图上添加统计结果

• 添加元素

• 格式调整ggpar

• 标签字体

• 对图形进行修饰annotate

• 改变坐标轴尺度xscale(), yscale()

• 改变作图区背景色

• 设置图区边线 border()

• 设置调整调色板

• 主题相关

• 旋转

• 图形组合 arrange

• 更改参数

• 添加元素

• 删除元素

• 差异表达数据 MA-plot

• 配对比较图 paired data plot

• 绘制表格

• 文字画

• 散点图 scatter plot

• 柱状图 ggbarplot()

• 点线图 ggline()

• 饼图(pie)

• 牛眼图(donutchart)

• 气泡图 bublle plot

• 误差图 errorplot

• 一维散点图 ggstrpchart()

• 箱线图 ggboxplot()

• 小提琴图 ggviolin

• 散点直方图（组合图形）ggscatterhist()

• Q-Q图（是否符合分布）ggqqplot()

• 直方图 gghistogram()

• 概率累积分布图

• 密度曲线图 ggdensity()

• 累积分布曲线图

• Cleveland点线图 ggdotchart()

• 点图 ggdotplot()

• 预加载包 loading

• ggpubr包简介

• 数据分布图 distribution

• 统计变换图 stat_transform

• 特殊数据图 special plot

• 图形调整

• 统计数据输出

• 元素输出

ggpubr包简介

ggplot2 (Hadley Wickham开发)是目前R语言数据可视化的主流。与R基础绘图系统相比，基于grid绘图系统的ggplot2已经在语法理解性上已经进步很多，但是通过ggplot2绘制用于学术杂志的图形，仍然需要较多的绘图函数（或者加载一些写好的模板代码）。为此Alboukadel Kassambara基于ggplot2、ggsci包开发了ggpubr用于绘制符合出版物要求的图形。该包封装了很多ggplot2的绘图函数，并且内嵌了ggsci中很多优秀的学术期刊配色方案，值得学习使用。

安装方法见https://rpkgs./ggpubr/index.html

• 目前也可以在CRAN上直接安装

• github install

预加载包 loading

需要预加载的包主要是：

• tidyverse

• ggpubr

数据分布图 distribution

分布特征图用于可视化数据的分布是否符合某种分布，将数据点离散分组，通过形状、颜色、大小等图形参数将数据分布差异显示出来。主要图形有：

• 一维散点图（数量少适用） strip chart

• 箱线图 boxplot

• 小提琴图 violin chart

• 散点直方图（组合图形）scatter histogram

• Q-Q图（是否符合分布）Q-Q chart

• 直方图 histo

• 密度曲线图 density chart

• Cleveland点线图 Cleveland’s Dot Plots

• 点图 Dot plot

一维散点图 ggstrpchart()

# x continousggstripchart(mpg,x="cyl",y="cty",  color="drv",  fill="drv",  palette="nejm",  shape=21,  size=2,  add=c("boxplot"),# one of “none”, “reg.line”, “loess”  add.params=list(color="black",fill="NA"),  error.plot="errorbar",  title="stripchart \n nejm palette",  # label="drv",  # font.label = list(size=14,face="italic",color="black"),  # label.select="r",  #repel=T,  #label.rectangle=T,  # jitter=0.2,  position=position_jitter(0.2),  ggtheme=theme_bw()  )
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21

#分类变量xggstripchart(mpg,x="class",y="cty",   color="drv",  #   fill="drv",  palette="nejm",  shape=21,  size=2,  add="mean_sd",# one of “none”, “reg.line”, “loess”  add.params=list(color="drv",fill="NA"), # error.plot="crossbar",  title="stripchart \n nejm palette",  #   label="drv",  # font.label = list(size=14,face="italic",color="black"),  # label.select="r",  #repel=T,  #label.rectangle=T,  jitter=0.5,  position=position_jitterdodge(),  ggtheme=theme_bw() )
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20

箱线图 ggboxplot()

ggboxplot(mpg, x = "drv", y = "cty",  color = "black", 
          fill="drv",  palette ="aaas", 
          width=0.2,  alpha=0.7,  # notch=T,  # remove=c("4")  select=c("f","r"),  order=c("r","f"),  add="jitter", #加散点  add.params = list(shape="drv",color="#00AFBB",alpha=0.6),  # label="fl",  # font.label="bold",  # label.select=list(top.up=10),  # repel=T,  # label.rectangle=T,  bxp.errorbar=T,  bxp.errorbar.width = 0.1,  xlab="drivers",  ylab=expression(speed~(m~s^-1)),  title="AAAS palette",  ggtheme=theme_pubr()  )
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24

小提琴图 ggviolin

ggviolin(mpg,x="class",y="hwy", # remove=c("4") select=c("compact","suv","midsize"), order=c("midsize","suv","compact"), color = "drv", 
         fill="NA", palette ="uchicago", 
         alpha=0.7, #透明度 linetype=1, trim=F, size=1, width=1, #violin width draw_quantiles = 0.5, #median value 
         add=c("jitter","mean"), #加散点 add.params = list(shape="drv",color="drv",alpha=0.6), error.plot = "errorbar", # label="fl", # font.label=c(size=11,color="black","bold"), # label.select=list(top.up=10), # repel=T, # label.rectangle=T, position=position_dodge(1), xlab="drivers", ylab=expression(speed~(m~s^-1)), title="uchicago palette", ggtheme=theme_pubr()  )
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28

散点直方图（组合图形）ggscatterhist()

ggscatterhist(mpg,x="hwy",y="cty",  color="drv",  fill="drv",  palette="material",  shape=21,  size=2,  linetype = "solid",  bins=30,  margin.plot="boxplot",  margin.params=list(color="drv",fill="NA"),  margin.ggtheme=theme_light(),  margin.space=F,  main.plot.size=2,  margin.plot.size=1,  title="material palette",  legend="right",  ggtheme=theme_cleveland()    
              )
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18

Q-Q图（是否符合分布）ggqqplot()

ggqqplot(mpg,x="displ", color="drv", palette="locuszoom", size=2, shape=21, add="qqline", add.params = list(linetype = "solid",size=1,linetype=1), conf.int=T, conf.int.level=0.95, title="locuszoon palette" )
1
2
3
4
5
6
7
8
9
10
11

直方图 gghistogram()

gghistogram(mpg,x="hwy",y="..count..",color="NA",fill="drv",palette = "igv",size=1,linetype=1,alpha=0.5,bins=30, #柱子个数binwidth=0.8,title="igv palette",xlab="hwy",ylab="Count",   # add="mean",add.params=list(size=1.5,linetype=1,color="drv"),rug=TRUE,add_density=T,   #   label="drv",# font.label = list(size=14,face="italic",color="black"), # label.select="r", #repel=T,  #label.rectangle=T,position=position_dodge(1))
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23

概率累积分布图

ggecdf(mpg,"hwy",   color="drv",palette="npg",   size=2,linetype=1,   ylab="Cumualtive density",   ggtheme=theme_classic2()   )
1
2
3
4
5
6

密度曲线图 ggdensity()

ggdensity(mpg,x="hwy",y="..density..",  color="black",  fill="drv",  palette = "simpsons",  size=1,  linetype=1,  alpha=0.5,  
          title="density plot \n simpsons palette",  xlab="hwy",  ylab="density",  add="median",  add.params=list(size=1,linetype=1,color="drv"),  rug=TRUE,#  label="drv", # font.label = list(size=14,face="italic",color="black"),#  label.select="r", # repel=T,#  label.rectangle=T,  ggtheme=theme_pubr()  
          )
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22

累积分布曲线图

ggecdf(mpg,"hwy",   color="drv",palette="npg",   size=2,linetype=1,   ylab="Cumualtive density",   ggtheme=theme_classic2()   )
1
2
3
4
5
6

Cleveland点线图 ggdotchart()

ggdotchart(mpg,x="class",y="hwy",   color="drv",   palette="jama",   shape=21,   size=4,   dot.size=2,   sorting="desc", #按照y排序  # add="segments",   add.params=list(size=1,linetype=2,color="red"),   x.text.col=F,   rotate=T,   title="Cleveland's Dot Plots \n jama palette",   xlab="cty",ylab="hwy",   label="model",   label.select="passat",   font.label=list(size=12,face="italic",color="black"),   label.rectangle=T,   repel=T,   position=position_dodge(0)  )
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20

点图 ggdotplot()

ggdotplot(mpg,x="class",y="hwy",  select=c("compact","midsize","suv"), 
          color="drv",  fill="NA", 
          palette="jama",   size=1,   binwidth = 0.5,   add=c("violin","mean_sd"),   add.params=list(size=1,color="drv"),  error.plot="errorbar", 
           title="dotplot \n jama palette",   xlab="cty",ylab="hwy",   label="model",   label.select="passat",   font.label=list(size=12,face="italic",color="black"),   label.rectangle=T,   repel=T,  ggtheme=theme_bw())
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19

统计变换图 stat_transform

统计变换图则是在统计变换基础上，进行数据可视化，一般不直接呈现原始数据，多求取数据平均值、方差，显示数据离散及组件差异。主要形式是根据数据是否离散或者连续，进行绘制：

• 散点图 连续x-y，还有分组变量

• 点线图 连续x-y

• 柱状图 x多为分类变量

• 饼图

• 牛眼图

• 气泡图

• 误差图

散点图 scatter plot

ggscatter(mpg,x="cty",y="hwy", color="drv", #边线颜色 # fill="drv", #填充色 palette = "NPG", # shape="fl", #点形状 size=3, #点大小 # point=F, #控制是否显示点  rug=T, #可以看y数据分布趋势 title="scatter plot \n NPG palette", xlab="city speed", ylab=expression(Highway~speed~(km~h^-1)), # facet.by="fl",#分面 add="reg.line", #fitting line (reg.line, loess) add.params=list(color="drv",size=1), conf.int=T, conf.int.level=0.95, # ellipse=T, # ellipse.level=0.95, # ellipse.type="t", # ellipse.alpha=0.3, # ellipse.border.remove=T, # mean.point=T, # mean.point.size=10, # star.plot=T, # star,plot.lty=1, # star.plot.lwd=2, # label="fl", # font.label="bold", # font.family="serif", # repel=T, # label.rectangle=T, cor.coef=T, # coef.coeff.args=list(method="pearson"),#参考stat_cor() cor.method="kendall", # cor.coef.coord=c(20.5,40.5) #标签位置 # cor.coef.size=10, # ggp=NULL, #添加到已有图 # show.legend.text=FALSE ggtheme=theme_pubr() )
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40

柱状图 ggbarplot()

ggbarplot(mpg, x="class", y="cty", 
          # merge=T,   select=c("compact","suv","midsize"),  # remove=c("suv","pickup"),  # order=c("midsize","compact"),#定义x排序  color="drv",#边线颜色  fill="NA",#映射填充色  palette ="bar plot \n lancet palette",#  size=1.05,  width=0.4, #bar宽度  add=c("mean_sd"),#添加误差线或其它  error.plot = c("upper_errorbar"),#误差线样式  add.params = list(color="drv",size=1.05),  label=T,  lab.col="navyblue",  lab.size=5,  lab.pos="in",  lab.vjust=1.5,  # lab.hjust=0.5,  lab.nb.digits=1,#小数位数  # sort.val="desc",#根据y值排序，但是x需要唯一  # sort.by.groups=F,  # top=2,#只显示最高的2个  position=position_dodge(),  orientation = "vertical",  title="lancet",  xlab="class",  ylab=expression(speed~(m~s^-1))  )
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29

点线图 ggline()

ggline(mpg,x="class",y="hwy",   color="drv",   palette="jco",   linetype="drv",   plot_type="b",#c(b,l,p)   size=1.2,   shape=19,   point.size=3,   # point.color="drv",   title="line chart \n JCO palette",   xlab="class",   ylab="highway speed",   # select=c(""),   # remove=c(),   # order=   add=c("mean_se"),   add.params=list(size=1.1),   error.plot = "errorbar",   # label="model",   #font.label="bold",   #label.select=list(top.up = 10, top.down = 4),   # repl=T,   #label.rectangle=T,   # show.line.label=T,   ggtheme=theme_pubr()   )
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26

饼图(pie)

饼图需要单一数据格式，ggpie不具有变换功能，因为ggplot2中需要极坐标转换。

aggregate(hwy~class,mpg,length)%>%ggpie(x="hwy",  label="hwy",  lab.pos="out",  lab.adjust=1,  lab.font=c(4,"bold","red"),  font.family="serif",  color="white",  fill="class",  palette = "ucscgb",  size=1.2,  title="pie chart \n ucscgb palette")
1
2
3
4
5
6
7
8
9
10
11
12

牛眼图(donutchart)

aggregate(hwy~class,mpg,length)%>%ggdonutchart(x="hwy", label="hwy", lab.pos="out", lab.adjust=0.5, lab.font=c(size=6,face="italic",color="black"), font.family="mono", color="white", fill="class", palette="donut chart \n rickandmorty", size=2 )
1
2
3
4
5
6
7
8
9
10
11
12

气泡图 bublle plot

气泡图用点图直径表示数据大小，可将三维数据二维化。需要格式化数据，ggballonplot没有统计变换功能。

aggregate(hwy~fl+class+drv,mpg,max)%>%
  ggballoonplot(x="fl",y="class",size="hwy",  size.range=c(1,10),  shape=21,  color="NA",  fill="drv",  #show.label=T,  #font.label=list(size=12,face="italic",color="black"),  rotate.x.text=F,  title="balloon plot",  ggtheme = theme_pubr()  )
1
2
3
4
5
6
7
8
9
10
11
12

误差图 errorplot

ggerrorplot(mpg,x="class",y="hwy",desc_stat="mean_sd", #核心参数color="drv",palette="gesa",size=1,width=0.2,title="error plot \n gesa palette",xlab="class",ylab="hwy",add=c("mean"),add.params=list(size=1,shape=19,color="drv",fill="white"),error.plot="errorbar",ci=0.95,position=position_dodge(0.8))
1
2
3
4
5
6
7
8
9
10
11
12
13
14

特殊数据图 special plot

差异表达数据 MA-plot

数据文件至少要包含：

• name: 基因名称

• baseMean: the mean expression of genes in the two groups.

• log2FoldChange: the log2 fold changes of group 2 compared to group 1

• padj: the adjusted p-value of the used statiscal test.

ggmaplot(diff_express[1:1000,], main = expression("Group 1" %->% "Group 2"), fdr = 0.05, fc = 2, size = 0.4, palette = c("#B31B21", "#1465AC", "darkgray"), genenames = as.vector(diff_express[1:1000,]$name), legend = "top", top = 20, #筛选显著20个基因 select.top.method="fc", #最大fold font.label = c("bold", 11), label.rectangle = TRUE, font.legend = "bold", font.main = "bold", ggtheme = ggplot2::theme_bw())
1
2
3
4
5
6
7
8
9
10
11

配对比较图 paired data plot

用于配对比较处理前后或者有无处理的差别。

ggpaired(mpg, cond1="cty",cond2="hwy", #  x="group",y="values"， #x为分组变量，y为观测值 id="model", color="condition", #  fill="condition", palette="npg", width=0.5, point.size=1, line.size=0.5, line.color = "gray", repel=T)
1
2
3
4
5
6
7
8
9
10
11
12
13

绘制表格

table1<-ggtexttable(mtcars[1:10,1:6], 
           theme= ttheme(# "minimal",base_size=10,base_colour="red" #非“default"下自定义无效, colnames.style=colnames_style(size=15,color="black",  face="bold",fill="grey80",  linewidth=4,linecolor="white"  ),rownames.style=rownames_style(color="blue",face="bold"),tbody.style=tbody_style(size=10,color="black",face="plain",fill=c("grey90","white"),#可色板linewidth=2,linecolor="white",#  hjust=0,x=0.1 #Left:hjust=0, x=0.1 Right:hjust=1, x=0.9)   ))table1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16

table1%>%table_cell_font(row=5,column = 3,face="bold.italic",size=15)%>%#修改表格单元格字体table_cell_bg(row=5,column = 3,fill="yellow",color="red",linewidth=8)#修改表格单元格颜色
1
2

文字画

ggparagraph(paste(rep("I have a dream",10),collapse=","),color="red",size=12,face="bold",family="serif",lineheight=1 #行距)
1
2
3
4

text_grob(paste(rep("I have a dream",10),collapse=","), 
          face = "italic", color = "steelblue",size=20,lineheight=1,family="serif",  rot=45) %>%as_ggplot()
1
2
3

图形调整

图形组合 arrange

p1<-ggscatter(mpg,"hwy","cty",color="drv",add="reg.line",shape=21)+theme_bw()p2<-ggbarplot(mpg,"class","hwy",fill="drv",width=0.5, 
          add="mean_sd",add.params = list(color="drv",size=1.2),  position=position_dodge(0.5))p3<-ggarrange(p1,p2,ncol=2,nrow=1,   widths=c(1,1), #图形宽度   align="h",   labels=list("(a)","(b)"),   vjust=2,hjust=-3,   legend=c(0.5,0.8),   common.legend = F)p3
1
2
3
4
5
6
7
8
9
10
11
12

更改参数

格式调整ggpar

ggarrange(p1,
  ggpar(p1,palette="jco",  # gradient.cols=get_palette("npg",5),#自定义色板  main="main",submain="submain",  # title="title",subtitle="subtitle", 与main一致  caption="caption",  xlab="xlab",ylab="ylab",  font.main=c(size=14,style="bold",color="red"),  font.x=c(size=14,style="bold",color="red"),  font.family="serif",  xlim=c(10,50),ylim=c(15,30),  xscale="none",yscale="none",format.scale = F,  legend="top",legend.title="drivers",font.legend=c(10,"bold","black"),  ticks=T,tickslab = T,  font.ticklab=c(10,"bold","black"),  #  font.xtickslab = c(10,"bold","black"),  #  font.ytickslab = c(10,"bold","black"),  x.text.angle=45,y.text.angle = 45,  xticks.by = 5,  # yticks.by = 5, #刻度线间隔数值every 5,不能同时xy调整  # rotate=T, #coord-flip()  # orientation = "vertical",#y倒叙  ggtheme=theme_bw()  )  )
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25

标签字体

p1+font(object="ylab",#title,caption,legend.text,xylab,xy.textsize=20,color="red",face="bold",family="serif") #element_text()参数均可
1
2
3

对图形进行修饰annotate

annotate_figure(p3,top="Figures Combination",bottom=text_grob("data source",color="red"),left="Left label",right="right label",fig.lab="Figure 1",fig.lab.pos="top.left",fig.lab.size=20,fig.lab.face = "italic")
1
2
3
4
5
6
7
8
9
10

改变坐标轴尺度xscale(), yscale()

ggarrange(p1,p1+xscale("log2",F))
1

改变作图区背景色

ggarrange(p1,p1+bgcolor("seagreen2"))
1

设置图区边线 border()

ggarrange(p1,
	p1+border(color="red",size=2,linetype=2))
1
2

设置调整调色板

# color_palette("palette_name") #设置color色板# fill_palette("palette_name") #设置fill色板ggarrange(p1,
	set_palette(p1,"npg"),
	change_palette(p1,"npg")) #更改色板
1
2
3
4
5
6

主题相关

ggarrange(p1,p1+clean_theme()) #去除坐标轴相关内容
1

ggarrange(p1,p1+theme_pubclean(base_size=15,flip=F)) #去除坐标轴线
1

ggarrange(p1,p1+labs_pubr()) #对标签进行格式化
1

ggarrange(p1,
				p1+theme_pubr(base_size=15,border=T,base_family = "serif"),
				legend="bottom")
1
2
3

ggarrange(p1,
		p1+theme_classic2(base_size=20))
1
2

旋转

#旋转rotatep2+
  rotate()+ #坐标轴反转，类似corrd_flip()
  # rotate_axis_text()
  rotate_x_text(angle=180,hjust=.5,vjust=1)+
  rotate_y_text(angle=45,hjust=0.5,vjust=0.5) #刻度线文字反转
1
2
3
4
5
6

添加元素

增加网格线add gridlines

p1+theme_pubr()+grids("xy",color="red",size=1,linetype=2)
1

增加分面

facet(p1,facet.by="cyl",  short.panel.labs = F,   # Allow long labels in panels  panel.labs.background = list(fill = "steelblue", color = "steelblue"))
1
2
3

增加文字标签

ggtext(mpg,"hwy","cty",   color="drv",palette="npg",face="bold",family="sans",   label="fl",repel=F,   label.select="r",label.rectangle = T,   show.legend = T,grouping.vars = "displ",   position=position_jitter(0.2),   ggp=p2, #add to p2, or NULL   ggtheme=theme_bw()   )
1
2
3
4
5
6
7
8
9

在已有图形上增加辅助统计量se,sd等

p4<-ggbarplot(mpg, x="class", y="cty",  add="mean_sd",  add.params=list(color="blue"),  width=1  )add_summary(p4,"mean_se",error.plot="errorbar",color="red",fill="NA",group=1,width=0.2,shape=19,size=2,ci=0.95,data=NULL,position=position_dodge(0.8))+theme_bw()
1
2
3
4
5
6
7
8
9
10
11
12

添加置信区间换图

p1+stat_chull(aes(color=drv,fill=drv),geom="polygon",alpha=0.2)
1

p1+stat_stars(aes(color=drv))
1

p1+stat_conf_ellipse(aes(color=drv,fill=drv),geom="polygon",alpha=0.2)
1

图上添加统计结果

p1+stat_cor(method="kendall",aes(color=drv),label.x=20) #添加相关分析结果及p-value
1

p1+stat_regline_equation(aes(group=drv,color=drv),formula=y~x) #添加回归线方程
1

p2+stat_compare_means(method="anova") #添加ANOVA显著性数值p-value
1

ggbarplot(mpg,"drv","hwy",add="mean",fill="drv")+
  compare_means(hwy~drv,mpg,paired=F,ref.group ="f")%>%mutate(y.position = c(45, 35))%>%stat_pvalue_manual(label = "p.adj", label.size=6)#手动添加p-value
1
2
3
4

添加元素

ggadd(p2,add=c("mean_sd","jitter"),color="drv",fill="drv",group=1,  width=1,shape=21,size=3,alpha=0.6,  jitter=0.2,binwidth=1,dotsize=1,error.plot="errorbar",  ci=0.95,position=position_dodge(1),  p_geom="")
1
2
3
4
5
6

删除元素

ggarrange(p1,p1+rremove(object="grid"))
1

#"xy.grid" for both x and y grids#"axis" x.axis#"xylab", "xy.title" or "axis.title" for both x and y axis labels#"xy.text" or "axis.text" for both x and y axis texts# "ticks" "x.ticks"#"legend.title" for the legend title#"legend" for the legend
1
2
3
4
5
6
7

统计数据输出

均值比较

compare_means(hwy~1,mpg,mu=100,method="wilcox.test") #wilcox检验或者t.test
1

## # A tibble: 1 x 8
##   .y.   group1 group2            p    p.adj p.format p.signif method  
##   <chr>  <dbl> <chr>         <dbl>    <dbl> <chr>    <chr>    <chr>   
## 1 hwy        1 null model 3.33e-40 3.30e-40 <2e-16   ****     Wilcoxon
1
2
3
4

compare_means(hwy~cyl,mpg,paired=F,method="t.test",ref.group = ".all.") #against all 比较
1

## # A tibble: 4 x 8
##   .y.   group1 group2        p    p.adj p.format p.signif method
##   <chr> <chr>  <chr>     <dbl>    <dbl> <chr>    <chr>    <chr> 
## 1 hwy   .all.  4      9.28e-15 2.80e-14 9.3e-15  ****     T-test
## 2 hwy   .all.  6      2.79e- 1 2.80e- 1 0.28     ns       T-test
## 3 hwy   .all.  8      3.43e-21 1.40e-20 < 2e-16  ****     T-test
## 4 hwy   .all.  5      5.21e-13 1.00e-12 5.2e-13  ****     T-test
1
2
3
4
5
6
7

compare_means(hwy~cyl,mpg,paired=F,method="wilcox.test",ref.group = NULL) #pairwise比较
1

## # A tibble: 6 x 8
##   .y.   group1 group2        p    p.adj p.format p.signif method  
##   <chr> <chr>  <chr>     <dbl>    <dbl> <chr>    <chr>    <chr>   
## 1 hwy   4      6      3.30e-16 1.60e-15 3.3e-16  ****     Wilcoxon
## 2 hwy   4      8      1.47e-24 8.80e-24 < 2e-16  ****     Wilcoxon
## 3 hwy   4      5      8.18e- 1 8.20e- 1 0.81763  ns       Wilcoxon
## 4 hwy   6      8      2.37e-13 9.50e-13 2.4e-13  ****     Wilcoxon
## 5 hwy   6      5      9.80e- 4 2.30e- 3 0.00098  ***      Wilcoxon
## 6 hwy   8      5      7.56e- 4 2.30e- 3 0.00076  ***      Wilcoxon
1
2
3
4
5
6
7
8
9

compare_means(hwy~cyl,mpg,paired=F,method="t.test",ref.group = "4") #against参考组pairwise比较
1

## # A tibble: 3 x 8
##   .y.   group1 group2        p    p.adj p.format p.signif method
##   <chr> <chr>  <chr>     <dbl>    <dbl> <chr>    <chr>    <chr> 
## 1 hwy   4      6      2.67e-16 5.30e-16 2.7e-16  ****     T-test
## 2 hwy   4      8      9.60e-38 2.90e-37 < 2e-16  ****     T-test
## 3 hwy   4      5      9.26e- 1 9.30e- 1 0.93     ns       T-test
1
2
3
4
5
6

compare_means(hwy~cyl,mpg,paired=F,method="anova",ref.group=NULL) #方差模型
1

## # A tibble: 1 x 6
##   .y.          p    p.adj p.format p.signif method
##   <chr>    <dbl>    <dbl> <chr>    <chr>    <chr> 
## 1 hwy   9.98e-44 1.00e-43 <2e-16   ****     Anova
1
2
3
4

compare_means(hwy~cyl,mpg,paired=F,method="kruskal.test") #非参检验
1

## # A tibble: 1 x 6
##   .y.          p    p.adj p.format p.signif method        
##   <chr>    <dbl>    <dbl> <chr>    <chr>    <chr>         
## 1 hwy   4.86e-32 4.90e-32 <2e-16   ****     Kruskal-Wallis
1
2
3
4

compare_means(hwy~cyl,mpg,paired=F,ref.group = "4",group.by="drv") #分组pairwise比较
1

## # A tibble: 5 x 9
##   drv   .y.   group1 group2        p    p.adj p.format p.signif method  
##   <chr> <chr> <chr>  <chr>     <dbl>    <dbl> <chr>    <chr>    <chr>   
## 1 f     hwy   4      6      5.87e-14 2.90e-13 5.9e-14  ****     Wilcoxon
## 2 f     hwy   4      8      1.01e- 1 2.00e- 1 0.10     ns       Wilcoxon
## 3 f     hwy   4      5      3.27e- 1 3.30e- 1 0.33     ns       Wilcoxon
## 4 4     hwy   4      6      2.03e- 7 6.10e- 7 2.0e-07  ****     Wilcoxon
## 5 4     hwy   4      8      1.50e-11 6.00e-11 1.5e-11  ****     Wilcoxon
1
2
3
4
5
6
7
8

compare_means(hwy~cyl,mpg,paired=F,ref.group = ".all.",group.by="drv") #分组pairwise比较
1

## # A tibble: 9 x 9
##   drv   .y.   group1 group2           p     p.adj p.format p.signif method 
##   <chr> <chr> <chr>  <chr>        <dbl>     <dbl> <chr>    <chr>    <chr>  
## 1 f     hwy   .all.  4          6.88e-5   4.10e-4 6.9e-05  ****     Wilcox~
## 2 f     hwy   .all.  6          8.55e-7   6.80e-6 8.6e-07  ****     Wilcox~
## 3 f     hwy   .all.  8          2.66e-1   1.00e+0 0.266    ns       Wilcox~
## 4 f     hwy   .all.  5          4.24e-1   1.00e+0 0.424    ns       Wilcox~
## 5 4     hwy   .all.  4          7.11e-8   6.40e-7 7.1e-08  ****     Wilcox~
## 6 4     hwy   .all.  6          3.48e-1   1.00e+0 0.348    ns       Wilcox~
## 7 4     hwy   .all.  8          3.34e-5   2.30e-4 3.3e-05  ****     Wilcox~
## 8 r     hwy   .all.  6          3.26e-2   1.60e-1 0.033    *        Wilcox~
## 9 r     hwy   .all.  8          4.56e-1   1.00e+0 0.456    ns       Wilcox~
1
2
3
4
5
6
7
8
9
10
11
12

compare_means(hwy~cyl,mpg,paired=F,ref.group = NULL,group.by="drv") 
1

## # A tibble: 10 x 9
##    drv   .y.   group1 group2        p    p.adj p.format p.signif method  
##    <chr> <chr> <chr>  <chr>     <dbl>    <dbl> <chr>    <chr>    <chr>   
##  1 f     hwy   4      6      5.87e-14 5.90e-13 5.9e-14  ****     Wilcoxon
##  2 f     hwy   4      8      1.01e- 1 4.00e- 1 0.1007   ns       Wilcoxon
##  3 f     hwy   4      5      3.27e- 1 7.10e- 1 0.3267   ns       Wilcoxon
##  4 f     hwy   6      8      7.45e- 1 7.40e- 1 0.7446   ns       Wilcoxon
##  5 f     hwy   6      5      1.40e- 3 8.40e- 3 0.0014   **       Wilcoxon
##  6 f     hwy   8      5      2.36e- 1 7.10e- 1 0.2357   ns       Wilcoxon
##  7 4     hwy   4      6      2.03e- 7 1.60e- 6 2.0e-07  ****     Wilcoxon
##  8 4     hwy   4      8      1.50e-11 1.40e-10 1.5e-11  ****     Wilcoxon
##  9 4     hwy   6      8      1.50e- 6 1.00e- 5 1.5e-06  ****     Wilcoxon
## 10 r     hwy   6      8      1.23e- 2 6.20e- 2 0.0123   *        Wilcoxon
1
2
3
4
5
6
7
8
9
10
11
12
13

分组描述统计desc_statby()

desc_statby(mpg,"hwy",c("drv","cyl"),ci=0.95)
1

## Warning in stats::qt(ci/2 + 0.5, data_sum$length - 1): 产生了NaNs
1

##   drv cyl length min max median     mean  iqr    mad        sd        se
## 1   4   4     23  20  28   25.0 24.60870 2.50 1.4826 2.5359081 0.5287734
## 2   4   6     32  17  25   19.0 19.50000 3.00 2.9652 2.8960485 0.5119539
## 3   4   8     48  12  23   17.0 16.35417 3.00 1.4826 2.2168560 0.3199756
## 4   f   4     58  24  44   29.0 30.46552 2.75 2.9652 4.0271688 0.5287932
## 5   f   5      4  28  29   29.0 28.75000 0.25 0.0000 0.5000000 0.2500000
## 6   f   6     43  17  29   26.0 25.06977 2.00 1.4826 2.1973507 0.3350928
## 7   f   8      1  25  25   25.0 25.00000 0.00 0.0000        NA        NA
## 8   r   6      4  24  26   25.5 25.25000 1.25 0.7413 0.9574271 0.4787136
## 9   r   8     21  15  26   20.0 20.19048 6.00 4.4478 3.4149531 0.7452039
##          ci range         cv        var
## 1 1.0966089     8 0.10304927  6.4308300
## 2 1.0441368     8 0.14851531  8.3870968
## 3 0.6437079    11 0.13555298  4.9144504
## 4 1.0588901    20 0.13218777 16.2180883
## 5 0.7956116     1 0.01739130  0.2500000
## 6 0.6762447    12 0.08764942  4.8283499
## 7       NaN     0         NA         NA
## 8 1.5234802     2 0.03791791  0.9166667
## 9 1.5544680    11 0.16913683 11.6619048
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20

元素输出

执行ggplot函数

p1+geom_exec(geom_point, data = mpg, #geom_** functions x = "hwy", y = "displ", size = "cyl",)
1
2

p1+geom_point(aes(hwy,displ,size=cyl),data=mpg) 
1

图例输出

#get legendp1%>%get_legend()%>%as_ggplot()
1
2

点线形式输出

#show point and line typeshow_point_shapes()
1
2

show_line_types()
1

调色板输出get palette

get_palette("npg", 3) #生成色板中3个过渡色
1

## [1] "#E64B35FF" "#4DBBD5FF" "#00A087FF"