R语言可视化10: 韦恩图/upset图 - VennDiagr
2023-04-26 本文已影响0人
小程的学习笔记
1. 使用
包绘制韦恩图
1.1 两个数据集
# 安装并加载所需的R包
# install.packages("VennDiagram")
library(VennDiagram)
# 创建测试数据
set1 <- sample(1:1000,300, replace = F) # replace = F是默认的,表示不放回抽样
set2 <- sample(1:1000,130, replace = F)
set3 <- sample(1:1000,300, replace = F)
set4 <- sample(1:1000,200, replace = F)
set5 <- sample(1:1000,300, replace = F)
s1 <- list(set1 = set1,
set2 = set2)
v1 <- venn.diagram(x = s1,
filename = NULL, # 直接给一个名称会自动保存文件到本地
# 输出的图形参数
# imagetype = "png", # 输出图片类型(tiff,png,svg)
# height = 1000, # 图片高度
# width = 1000, # 图片宽度
# resolution = 300, # 图片分辨率
scaled = T, # 根据比例显示大小
alpha=c(0.8, 0.8), # 设置每个区块的透明度
## 下面是除了标题外，图形其他元素的设置参数
# 图形元素设置：圈
lwd = 1, # 圆圈线条的粗细：1 2 3 4 5 6
lty = 1, # 圆圈线条的类型：1为实线，2为虚线，blank为无线条
col = c("black","red"), # 圆圈线条颜色
fill = c("#0073C2FF", "#EFC000FF"), # 圆圈颜色
# 图形元素设置：数字
cex = 1, # 数字大小
fontface = "bold", # 加粗
fonrfamily = "sans", # 数字字体
# 图形元素设置：标签即(category)
cat.cex = 1, # 标签字体大小
cat.col = "black", # 标签字体色
cat.fontface = "bold", # 加粗
cat.default.pos = "outer", # 标签内外位置, 在圆圈内还是圆圈外，outer 内 text 外
cat.pos = c(0, 0), # 标签旋转位置，用圆的度数
cat.dist = c(0.05,0.03), # 标签离圆圈位置,离圆的距离，如果标签与圆圈重叠，可以调整这个参数
cat.fontfamily = "sans", # 标签字体
)
cowplot::plot_grid(v1)
VennDiagram-1
1.2 多个数据集（此处以5个为示例）
s2 <- list(
set1 = set1,
set2 = set2,
set3 = set3,
set4 = set4,
set5 = set5
)
v2 <- venn.diagram(x = s2, filename = NULL,
col = "transparent",
fill = c("dodgerblue", "goldenrod1", "darkorange1", "seagreen3", "orchid3"),
label.col = c("dodgerblue", "goldenrod1","darkorange1","seagreen3", "orchid3","white", "white",
"white","white","white","white","white","white", "white","white","white","white",
"white","white","white", "white", "white", "white", "white", "white","white",
"white","white", "white", "white", "black"),
fontface = "bold",
cat.col = c(cat.col = c("darkblue", "darkgreen", "orange", "grey50", "purple")),
cat.dist = c(0.2, 0.2, 0.18, 0.18, 0.2),
alpha = 0.50,
cex = 1,
cat.cex = 1,
margin = 0.05
)
cowplot::plot_grid(v2)
VennDiagram-2
1.3 交集元素的提取
# VennDiagram包中的函数get.venn.partitions()提供了此这个功能
# 以上述5个分组为例，组间交集元素获得
inter <- get.venn.partitions(s2)
head(inter)
## set1 set2 set3 set4 set5 ..set.. ..values.. ..count..
## 1 TRUE TRUE TRUE TRUE TRUE set1∩set2∩set3∩set4∩set5 822, 588 2
## 2 FALSE TRUE TRUE TRUE TRUE (set2∩set3∩set4∩set5)∖(set1) 406 1
## 3 TRUE FALSE TRUE TRUE TRUE (set1∩set3∩set4∩set5)∖(set2) 442, 104 2
## 4 FALSE FALSE TRUE TRUE TRUE (set3∩set4∩set5)∖(set1∪set2) 366, 715, 379, 414, 30, 308, 398, 322, 359, 825, 708, 458 12
## 5 TRUE TRUE FALSE TRUE TRUE (set1∩set2∩set4∩set5)∖(set3) 615, 541 2
## 6 FALSE TRUE FALSE TRUE TRUE (set2∩set4∩set5)∖(set1∪set3) 934, 84, 75, 655 4
★ 5个数据集是VennDiagram包的上限
2. 使用
包绘制韦恩图
# 安装并加载所需的R包
# install.packages("ggVennDiagram")
library(ggplot2)
library(ggVennDiagram)
# ggVennDiagram提供了不同的形状以供选择，默认情况下，只使用最合适的形状，但也可自行指定形状
plot_shapes()
ggVennDiagram-1
2.1 三个数据集
x1 <- list(
set1 = set1,
set2 = set2,
set3 = set3
)
# method1
ggVennDiagram(x1, category.names = c("A", "B", "C"), # 设定样本名称
label = "both", # 可选："both", "count", "percent", "none"
label_color = "black",
label_alpha = 0, # 去除文字标签底色
edge_lty = "dashed", # 圆圈线条虚线
edge_size = 1) +
scale_fill_gradient(low = "white", high = "#b9292b", name = "gene count")
# method2
# 构建维恩对象
venn <- Venn(x1)
data <- process_data(venn, shape_id == "301")
ggplot() +
geom_sf(aes(fill = count),
data = venn_region(data)) +
geom_sf(color="grey",
size = 1,
data = venn_setedge(data),
show.legend = FALSE) +
scale_fill_gradient(low ="white", high = "#b9292b", name = "gene count")+
geom_sf_text(aes(label = name),
data = venn_setlabel(data),
size = 8) +
geom_sf_label(aes(label = count),
data = venn_region(data),
size = 4) +
theme_void()
ggVennDiagram-2
2.2 多个数据集（此处以5个为示例）
# 不添加过多的填充颜色，可在Ai中进行后期调整
library(ggsci)
ggVennDiagram(x2, , label_alpha = 0, label = "none",
edge_size = 0.5,
# show_intersect = TRUE # 用交互的方式（plotly）查看每个子集中的基因
) +
scale_color_lancet() + # R包"ggsci"，柳叶刀期刊色标
scale_fill_gradient(low = "gray100", high = "gray95", guide = "none")
# 自定义颜色；
color1 <- alpha("#f8766d", 0.9)
ggVennDiagram(x2, label_alpha = 0, label_size = 3,
# edge_size = 0.5, label ="count", # 隐藏百分比, 默认"both"
# show_intersect = TRUE # 用交互的方式（plotly）查看每个子集中的基因
) +
scale_color_brewer(palette = "Paired") +
scale_fill_gradient(low = "white", high = color1,
guide="none" # 去除图例
)
ggVennDiagram-3
★ 支持1-7维的韦恩图绘制
★ 是ggplot2的拓展包，因此支持ggplot2的其他语法设置
★ show_intersect = T时，可输出为交互式html，此时可点击数值显示源数据
3. 使用
包绘制upset图
UpsetR包，经常用于大于5个样本的“韦恩图”
# 安装并加载所需的R包
# install.packages("UpSetR")
# install.packages("RColorBrewer")
# 安装一个数据集
install.packages("ggplot2movies")
library(UpSetR)
library(RColorBrewer)
library(ggplot2)
# 使用的来自IMDB中的电影数据
movies <- as.data.frame(ggplot2movies::movies)
head(movies)
## title year length budget rating votes r1 r2 r3 r4 r5 r6 r7 r8 r9 r10 mpaa Action Animation Comedy Drama Documentary Romance Short
## 1 $ 1971 121 NA 6.4 348 4.5 4.5 4.5 4.5 14.5 24.5 24.5 14.5 4.5 4.5 0 0 1 1 0 0 0
## 2 $1000 a Touchdown 1939 71 NA 6.0 20 0.0 14.5 4.5 24.5 14.5 14.5 14.5 4.5 4.5 14.5 0 0 1 0 0 0 0
## 3 $21 a Day Once a Month 1941 7 NA 8.2 5 0.0 0.0 0.0 0.0 0.0 24.5 0.0 44.5 24.5 24.5 0 1 0 0 0 0 1
## 4 $40,000 1996 70 NA 8.2 6 14.5 0.0 0.0 0.0 0.0 0.0 0.0 0.0 34.5 45.5 0 0 1 0 0 0 0
## 5 $50,000 Climax Show, The 1975 71 NA 3.4 17 24.5 4.5 0.0 14.5 14.5 4.5 0.0 0.0 0.0 24.5 0 0 0 0 0 0 0
## 6 $pent 2000 91 NA 4.3 45 4.5 4.5 4.5 14.5 14.5 14.5 4.5 4.5 14.5 14.5 0 0 0 1 0 0 0
# 调整与美化后的集合图#
upset(fromList(movies),
nsets = length(movies), # 显示数据集的所有数据, nsets = 数值调整可视化数据集数量
nintersects = 15, # 显示前多少个
sets = c("title","length","budget","votes","year"), # keep.order = TRUE, # 指定集合或用keep.order = TRUE保持集合按输入的顺序排序
number.angles = 0, # 交互集合柱状图的柱标倾角
point.size = 4, # 图中点的大小
line.size = 1, # 图中连接线粗细
mainbar.y.label = "Intersection size", # y轴的标签
main.bar.color = 'black', # y轴柱状图颜色
matrix.color = "black", # x轴点的颜色
sets.x.label = "Set size", # x轴的标签
sets.bar.color=brewer.pal(5,"Set1"), # x轴柱状图的颜色; Set1中只有9个颜色，Set3中有12个颜色，Paired中有12个颜色
mb.ratio = c(0.7, 0.3), # bar plot和matrix plot图形高度的占比
order.by = "freq", # y轴矩阵排序,如"freq"频率，"degree"程度
text.scale = c(1.5, 1.5, 1.5, 1.5, 1.5, 1), # 6个参数intersection size title（y标题大小）,intersection size tick labels（y刻度标签大小）, set size title（set标题大小）, set size tick labels（set刻度标签大小）, set names（set 分类标签大小）, numbers above bars（柱数字大小）的设置
shade.color = "#12507B", # 图中阴影部分的颜色
queries=list(list(query = intersects, params = list("votes"), color = "purple", active = T), # 设置自己想要展示的特定组的交集，通过queries参数进行设置，需要展示几个关注组合的颜色，就展示几个
list(query = intersects, params = list("votes","length"), color = "orange", active = T))
)
upsetR-1
★ 不支持ggplot语法
4. 使用
包绘制upset图
4.1 基本用法
# 安装并加载所需的R包
# install.packages('ComplexUpset')
# if(!require(devtools)) install.packages("devtools")
# devtools::install_github("krassowski/complex-upset")
library(ggplot2)
library(ComplexUpset)
movies = as.data.frame(ggplot2movies::movies)
# 第18-24列是电影类型(用0,1矩阵表示)
genres <- colnames(movies)[18:24]
genres
## [1] "Action" "Animation" "Comedy" "Drama" "Documentary" "Romance" "Short"
# 把mpaa这一列中的空值变成NA，然后为了方便演示去掉缺失值
movies[movies$mpaa == "", "mpaa"] <- NA
movies <- na.omit(movies)
upset(movies, genres,
name='genre', # 底部的标签
width_ratio = 0.2, # 左侧柱状图的宽度
height_ratio = 0.3, # 下图部分比例
min_size = 5, # 显示的最小集合的大小
min_degree = 2, # 最小等级，即显示最少几个数据集的集合
n_intersections = 15,
wrap = TRUE, set_sizes = FALSE
)
ComplexUpset-1
4.2 添加组件（annotations）
# 三种方法添加多个注释组件
upset(
movies,
genres,
annotations = list(
# 方法1-使用list：添加length这一列数据
'Length'= list(
aes = aes(x = intersection, y = length),
geom = geom_boxplot(na.rm = TRUE)
),
# 方法2-使用ggplot2：添加rating这一列数据
'Rating'=(
# aes(x=intersection) 是默认提供的，可以跳过
ggplot(mapping = aes(y = rating))
+ geom_jitter(aes(color = log10(votes)), na.rm = TRUE)
+ geom_violin(alpha = 0.5, na.rm = TRUE)
),
# 方法3：使用内置的 upset_annotate() 函数
'Budget'=upset_annotate('budget', geom_boxplot(na.rm=TRUE))
),
min_size = 10,
width_ratio = 0.1
)
# 使用条形图来展示分类变量比例的差异
upset(
movies,
genres,
annotations = list(
'MPAA Rating'= (
ggplot(mapping = aes(fill = mpaa))
+ geom_bar(stat = 'count', position = 'fill')
+ scale_y_continuous(labels = scales::percent_format())
+ scale_fill_manual(values = c(
'R' = '#E41A1C', 'PG' = '#377EB8',
'PG-13' = '#4DAF4A', 'NC-17' = '#FF7F00'
))
+ ylab('MPAA Rating')
)
),
width_ratio = 0.1
)
ComplexUpset-2
4.3 区域选择模式
ComplexUpset提供
定义相应维恩图上的感兴趣区域(以A、B、C三个数据集为例），自定义时，可用intersection_size()进行相应地调整
：
upset(
upset(
movies, genres,
mode = 'inclusive_intersection',
annotations = list(
# # 这里如果不指定就会使用上面设置好的模式)
'Length (inclusive intersection)' = (
ggplot(mapping = aes(y = length))
+ geom_jitter(alpha = 0.2, na.rm = TRUE)
),
'Length (exclusive intersection)' = (
ggplot(mapping = aes(y = length))
+ geom_jitter(alpha = 0.2, na.rm = TRUE)
+ upset_mode('exclusive_intersection')
),
'Length (inclusive union)' = (
ggplot(mapping = aes(y = length))
+ geom_jitter(alpha = 0.2, na.rm = TRUE)
+ upset_mode('inclusive_union')
)
),
min_size = 10,
width_ratio = 0.1
)
# 增加颜色映射
library(ggsci)
upset(movies, genres,
min_size = 10, width_ratio = 0.1,
# 调整intersection size
base_annotations = list(
"intersection size" = intersection_size(
counts = F, # 不显示个数
mapping = aes(fill = "bars_color")
)
+ scale_fill_manual(values = c("bars_color" = "skyblue"), guide = "none") # 使用单一颜色
)
)
upset(movies, genres,
min_size = 10, width_ratio = 0.1,
# 调整intersection size
base_annotations = list(
"intersection size" = intersection_size(
counts = F, # 不显示个数
mapping = aes(fill = mpaa)
)
+ scale_fill_lancet() # 使用ggsci包的lancet配色
)
)
ComplexUpset-3
5. 使用
包，韦恩图+韦恩条形图+韦恩饼图+upset图
5.1 不同布局的图形
# 安装并加载所需的R包
# if (!requireNamespace("BiocManager"))
# install.packages("BiocManager")
# BiocManager::install("VennDetail")
library(VennDetail)
# 创建测试数据
A <- sample(1:1000, 400, replace = FALSE)
B <- sample(1:1000, 600, replace = FALSE)
C <- sample(1:1000, 350, replace = FALSE)
D <- sample(1:1000, 550, replace = FALSE)
E <- sample(1:1000, 450, replace = FALSE)
venn <- venndetail(list(A = A, B = B, C= C, D = D, E = E))
detail(venn)
# 韦恩图（默认）
plot(venn)
# 韦恩饼图
plot(venn, type = "vennpie")
vennpie(venn,
min = 4 # 显示集合至少包含来自四个数据集的元素
# any = 1, revcolor = "lightgrey" # 突出显示唯一或共享子集
)
# 韦恩条形图
dplot(venn, order = TRUE, textsize = 4)
# upset图
plot(venn, type = "upset")
VennDetail-1
5.2 提取子集及可用注释
## 列出子集名称
detail(venn)
## Shared B_C_D_E A_C_D_E C_D_E A_B_D_E B_D_E A_D_E D_E A_B_C_E B_C_E A_C_E C_E A_B_E B_E
## 15 27 14 23 51 59 29 38 17 22 11 14 29 50
## A_E E A_B_C_D B_C_D A_C_D C_D A_B_D B_D A_D D A_B_C B_C A_C C
## 19 32 28 43 7 27 34 61 32 62 30 37 14 21
## A_B B A
## 49 48 21
head(getSet(venn, subset = c("Shared", "A_C_D_E")), 10)
## Subset Detail
## 1 Shared 522
## 2 Shared 413
## 3 Shared 362
## 4 Shared 415
## 5 Shared 789
## 6 Shared 984
## 7 Shared 712
## 8 Shared 719
## 9 Shared 114
## 10 Shared 666
head(result(venn, wide = TRUE))
## Detail A B C D E SharedSets
## 10 522 1 1 1 1 1 5
## 52 413 1 1 1 1 1 5
## 116 362 1 1 1 1 1 5
## 136 415 1 1 1 1 1 5
## 177 789 1 1 1 1 1 5
## 185 984 1 1 1 1 1 5
参考：
