R - Combine cor.mtest and p.adjust
Is there a way to account for multiple p-value comparisons (like p.adjust) using this cor.mtest function? Code from http://cran.r-project.org/web/packages/corrplot/vignettes/corrplot-intro.html
Thank!
cors<-cor(rel_tnum_data)
cor.mtest <- function(mat, conf.level = 0.95) {
mat <- as.matrix(mat)
n <- ncol(mat)
p.mat <- lowCI.mat <- uppCI.mat <- matrix(NA, n, n)
diag(p.mat) <- 0
diag(lowCI.mat) <- diag(uppCI.mat) <- 1
for (i in 1:(n - 1)) {
for (j in (i + 1):n) {
tmp <- cor.test(mat[, i], mat[, j], conf.level = conf.level)
p.mat[i, j] <- p.mat[j, i] <- tmp$p.value
lowCI.mat[i, j] <- lowCI.mat[j, i] <- tmp$conf.int[1]
uppCI.mat[i, j] <- uppCI.mat[j, i] <- tmp$conf.int[2]
}
}
return(list(p.mat, lowCI.mat, uppCI.mat))
}
res1 <- cor.mtest(rel_tnum_data, 0.95)
res2 <- cor.mtest(rel_tnum_data, 0.99)
corrplot(cors, p.mat = res1[[1]], sig.level=0.05, insig="blank", cl.align="r", tl.cex=0.6, order="hclust", type="lower", tl.srt=60, cl.ratio=0.1)
+3
source to share
1 answer
You can vectorize your matrix of p values โโ(like res1) that was returned by the function cor.mtest
. Subsequently, the function p.adjust
will help you get the adjusted p values.
pAdj <- p.adjust(c(res1[[1]]), method = "BH")
Then create the original matrix with the p-values โโset.
resAdj <- matrix(pAdj, ncol = dim(res1[[1]])[1])
Then you corrplot
should be ready to create your graph with the p-values โโset.
corrplot(cors, p.mat = resAdj, sig.level=0.05, insig="blank", cl.align="r", tl.cex=0.6, order="hclust", type="lower", tl.srt=60, cl.ratio=0.1)
Please note that p.mat = resAdj
.
+1
source to share