Xueyd's Blog

0%

R语言绘制韦恩图和upset图

发表于 更新于 分类于
本文字数： 447 阅读时长 ≈ 1 分钟

本文总结一下：韦恩图、upset图。

1. 韦恩图

1.1 ggvenn

利用ggvenn绘制韦恩图，ggvenn的输入是一个list。

1
2
3
library(ggvenn)

a <- list(A=c(1,2,3,5,7),B=c(2,3,4,1,6),C=c(2,3,7,6,8),D=c(1,2,5,6,3))

ggvenn默认将所有的set都绘制韦恩图，也可以通过set名字指定数据集绘制韦恩图。

1
ggvenn(a)

alt 图标

1
ggvenn(a,c("A","B"))

alt 图标

1
ggvenn(a,c("A","B","C"))

alt 图标
美化一下：

1
2
3
4
5
6
list(A=a[[1]],B=a[[2]],C=a[[3]],D=a[[4]]) %>% 
  ggvenn(show_percentage = T,show_elements = F,label_sep = ",",
         digits = 1,stroke_color = "white",
         fill_color = c("#E41A1C", "#1E90FF", 
                        "#FF8C00","#4DAF4A"),
         set_name_color = c("#E41A1C", "#1E90FF","#FF8C00","#4DAF4A"))

alt 图标

1.2 VennDiagram

利用VennDiagram包绘制韦恩图,会保存到文件里。

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
venn.plot <- venn.diagram(
  x = list(A=c(1,2,3,5,7),B=c(2,3,4,1,6),C=c(2,3,7,6,8),D=c(1,2,5,6,3)),
  filename = "Venn.png", imagetype = "png",
  lwd = 3, lty="dotted",
  fill = c("#E41A1C", "#1E90FF","#FF8C00","#4DAF4A"),
  alpha = 0.6,
  label.col = "white",
  cex = 1.5,
  fontfamily = "serif",
  fontface = "bold",
  cat.col = c("#E41A1C", "#1E90FF","#FF8C00","#4DAF4A"),
  cat.cex = 2,
  cat.fontfamily = "serif",
  cat.fontface = "bold",
  margin = 0.05)

alt 图标

1.3 VennDiagram提取交集元素

VennDiagram包中的函数get.venn.partitions()可以实现这个功能：

1
2
3
4
5
venn_list <- list(A=c(1,2,3,5,7),B=c(2,3,4,1,6),C=c(2,3,7,6,8),D=c(1,2,5,6,3))

inter <- get.venn.partitions(venn_list)
for (i in 1:nrow(inter)) inter[i,'values'] <- paste(inter[[i,'..values..']], collapse = ', ')
write.table(inter[-c(5, 6)], 'venn4_inter.txt', row.names = FALSE, sep = '\t', quote = FALSE)

venn4_inter.txt结果如下：

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
A       B       C       D       ..count..       values
TRUE    TRUE    TRUE    TRUE    2       2, 3
FALSE   TRUE    TRUE    TRUE    1       6
TRUE    FALSE   TRUE    TRUE    0
FALSE   FALSE   TRUE    TRUE    0
TRUE    TRUE    FALSE   TRUE    1       1
FALSE   TRUE    FALSE   TRUE    0
TRUE    FALSE   FALSE   TRUE    1       5
FALSE   FALSE   FALSE   TRUE    0
TRUE    TRUE    TRUE    FALSE   0
FALSE   TRUE    TRUE    FALSE   0
TRUE    FALSE   TRUE    FALSE   1       7
FALSE   FALSE   TRUE    FALSE   1       8
TRUE    TRUE    FALSE   FALSE   0
FALSE   TRUE    FALSE   FALSE   1       4
TRUE    FALSE   FALSE   FALSE   0

2. upset图

UpSetR 提供了两个转换函数 fromList 和 fromExpression 用于格式化数据。

1
2
3
venn_list <- list(A=c(1,2,3,5,7,9,10),B=c(2,3,4,1,6),C=c(2,3,7,8),D=c(1,2,5,6,3,5,8,10,14,15))

upset(fromList(venn_list))

等价于：

1
2
b <- fromList(venn_list)
upset(b)

alt 图标

order.by默认升序排列，降序排列加decreasing=FALSE:

1
upset(fromList(venn_list), order.by = "freq")

alt 图标

1
upset(fromList(venn_list), order.by = "freq",descreasing = FALSE)

alt 图标

绘制部分集合：

1
upset(fromList(venn_list), nsets=2, order.by = "freq")

alt 图标

美化：

1
2
3
4
5
6
7
upset(fromList(venn_list), nsets = 6, 
      number.angles = 30, # 柱子上方数字倾斜角度
      point.size = 3.5,  # 点的大小
      line.size = 2,  # 线的大小
      mainbar.y.label = "Intersections", 
      sets.x.label = "set", 
      text.scale = c(1.3, 1.3, 1, 1, 2, 1))

text.scale 参数值的顺序为：

  • 柱状图的轴标签和刻度
  • 条形图的轴标签和刻度
  • 集合名称
  • 柱子上方表示交集大小的数值

alt 图标

参考资料：

  1. https://zhuanlan.zhihu.com/p/370210775