image
基于ggplot2包以及corrplot包的相关矩阵可视化包ggcorrplot，ggcorrplot包提供对相关矩阵重排序以及在相关图中展示显著性水平的方法，同时也能计算相关性p-value
<a id="more"></a>

计算相关矩阵(cor()计算结果不提供p-value)

library(ggcorrplot)
data("mtcars")
corr <- round(cor(mtcars), 1)
head(corr[, 1:6])

image

用ggcorrplot包提供的函数cor_pmat()

p.mat <- cor_pmat(mtcars)

head(p.mat[, 1:4])

image

可视化相关性矩阵

ggcorrplot(corr)#method默认为square

image

方法为circle

ggcorrplot(corr, method = "circle")

image

重排矩阵，使用分等级聚类

ggcorrplot(corr, hc.order = TRUE, outline.color = "white")

image

控制矩阵形状

ggcorrplot(corr, hc.order = TRUE, type = "lower", outline.color = "white")#下三角形

image

上三角形

ggcorrplot(corr, hc.order = TRUE, type = "upper", outline.color = "white")

image

更改颜色以及主题

ggcorrplot(corr, hc.order = TRUE, type = "lower", outline.color = "white", 

ggtheme = ggplot2::theme_gray, colors = c("#6D9EC1", "white", "#E46726"))

image

添加相关系数

ggcorrplot(corr, hc.order = TRUE, type = "lower", lab = TRUE)

image

增加显著性水平，不显著的话就不添加了

ggcorrplot(corr, hc.order = TRUE, type = "lower", p.mat = p.mat)

image

将不显著的色块设置成空白

ggcorrplot(corr, p.mat = p.mat, hc.order=TRUE, type = "lower", insig = "blank")

image

转载https://ytlogos.github.io/2017/06/25/R%E8%AF%AD%E8%A8%80%E5%8F%AF%E8%A7%86%E5%8C%96%E5%AD%A6%E4%B9%A0%E7%AC%94%E8%AE%B0%E4%B9%8B%E7%9B%B8%E5%85%B3%E7%9F%A9%E9%98%B5%E5%8F%AF%E8%A7%86%E5%8C%96%E5%8C%85ggcorrplot/