相关矩阵可视化包ggcorrplot

       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-valuelibrary(ggcorrplot)#计算相关矩阵(cor()计算结果不提供p-value)data("mtcars")corr <- round(cor(mtcars), 1)head(corr[, 1:6])


#用ggcorrplot包提供的函数cor_pmat()p.mat <- cor_pmat(mtcars)head(p.mat[, 1:4])


可视化相关性矩阵ggcorrplot(corr)#method默认为square#方法为circleggcorrplot(corr, method = "circle")


#重排矩阵，使用分等级聚类ggcorrplot(corr, hc.order = TRUE, outline.color = "white")


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


#上三角形ggcorrplot(corr, hc.order = TRUE, type = "upper", outline.color = "white")


#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"))


#添加相关系数ggcorrplot(corr, hc.order = TRUE, type = "lower", lab = TRUE)


#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)


#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")