基于R的linkET包qcorrplot可视化Mantel test相关性网络热图分析correlation heatmap

写在前面

需求是对瘤胃宏基因组结果鉴定到的差异菌株与表观指标、瘤胃代谢组、血清代谢组、牛奶代谢组中有差异的部分进行关联分析，效果图如下：

数据准备

逗号分隔的csv格式文件，两个表格，一个是每个样本对应的表观指标数据，另一个是每个样本对应的菌群丰度，我这里用的是genus水平

• 需要关联的表观数据rumen.csv

• 不同样本的菌群丰度genus.csv

R包linkET可视化

• 装包

install.pakages("linkET")
library(linkET)
1
2

如果报错R版本有问题装不上（我的4.3.1版本R出现了这个报错）请尝试：

install.packages("devtools")
devtools::install_github("Hy4m/linkET", force = TRUE)
packageVersion("linkET")
1
2
3

• 读取数据

library(ggplot2)
rumen <- read.csv("rumen.csv",sep=",",row.name=1,stringsAsFactors = FALSE,check.names = FALSE)
genus <- read.csv("genus.csv",sep=",",row.name=1,stringsAsFactors = FALSE,check.names = FALSE)
#如果报错row.names重复错误请检查数据格式是否为csv
1
2
3
4

• rumen.csv组内相关系数

matrix_data(list(rumen = rumen)) %>% 
  as_md_tbl()
correlate(rumen) %>% 
  as_matrix_data()
correlate(rumen) %>% 
  as_md_tbl()

correlate(rumen) %>% 
  as_md_tbl() %>% 
  qcorrplot() +
  geom_square()

#如果对“%>%”功能报错，装具有此功能的包即可，比如dplyr

library(vegan)
correlate(rumen, genus, method = "spearman") %>% 
  qcorrplot() +
  geom_square() +
  geom_mark(sep = '\n',size = 3, sig_level = c(0.05, 0.01, 0.001),
            sig_thres = 0.05, color = 'white') + #添加显著性和相关性值
  scale_fill_gradientn(colours = RColorBrewer::brewer.pal(11, "RdBu"))
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21

• 两个表格进行关联生成相关性矩阵图，带显著性标记

library(vegan)
correlate(rumen, genus, method = "spearman") %>% 
  qcorrplot() +
  geom_square() +
  geom_mark(sep = '\n',size = 3, sig_level = c(0.05, 0.01, 0.001),
            sig_thres = 0.05, color = 'white') + #添加显著性和相关性值
  scale_fill_gradientn(colours = RColorBrewer::brewer.pal(11, "RdBu"))
1
2
3
4
5
6
7

• 加工可视化

library(dplyr)
mantel <- mantel_test(rumen, genus,
                      spec_select = list(Milk_yeild=1,Milk_fat=2,Urea_Nitrogen=3,Butyric_acid=4,Valeric_acid=5,BUN=6,
                                         T_AOC=7,SOD=8,MDA=9,IgA=10,IgG=11))%>% 
  mutate(rd = cut(r, breaks = c(-Inf,  0.5, Inf),
                  labels = c("< 0.5", ">= 0.5")),
         pd = cut(p, breaks = c(-Inf, 0.01, 0.05, Inf),
                  labels = c("< 0.01", "0.01 - 0.05", ">= 0.05")))

qcorrplot(correlate(genus), type = "lower", diag = FALSE) +
  geom_square() +geom_mark(sep = '\n',size = 1.8, sig_level = c(0.05, 0.01, 0.001),
    sig_thres = 0.05,color="white") +
  geom_couple(aes(colour = pd, size = rd), 
              data = mantel, 
              curvature = nice_curvature()) +
  scale_fill_gradientn(colours = RColorBrewer::brewer.pal(11, "RdBu")) +
  scale_size_manual(values = c(0.5, 1, 2)) +
  scale_colour_manual(values = color_pal(3)) +
  guides(size = guide_legend(title = "Mantel's r",
                             override.aes = list(color = "black"), 
                             order = 2),
         colour = guide_legend(title = "Mantel's p", 
                               override.aes = list(size = 3), 
                               order = 1),
         fill = guide_colorbar(title = "Pearson's r", order = 3))
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

• 不显著的灰色连接线部分也可以去掉让画面更干净。其余细节去AI加工即可。