R/Plot.pca.pvalues.R

In muma: Metabolomics Univariate and Multivariate Analysis

Plot.pca.pvalues <-
function(pcx, pcy, scaling) {
 loadi = paste(getwd(), "/PCA_Data_", scaling, "/PCA_LoadingsMatrix.csv", sep="")
 Loading <- read.csv(loadi, sep=",", header=TRUE)
 Loading.x <- Loading[,2:ncol(Loading)]
 rownames(Loading.x) <- Loading[,1]
 load.xlab = paste("Loading PC",pcx)
 load.ylab = paste("Loading PC",pcy)
 lim.load = c()
 Max.pc1 = 1.1*(max(Loading.x[,pcx]))
 Min.pc1 = 1.1*(min(Loading.x[,pcy]))
 Mpc1=c(Min.pc1*2,Max.pc1*2)
 Max.pc2 = 1.1*(max(Loading.x[,pcx]))
 Min.pc2 = 1.1*(min(Loading.x[,pcy]))
 Mpc2=c(Min.pc2*2,Max.pc2*2)
 colcool = "Colors_Pvalues"
 pwdcol = paste(getwd(), "/Univariate/Pvalues/", colcool, sep="")
 col.pv = read.csv(pwdcol, header=TRUE)
 col.pv = matrix(col.pv[,-1], ncol=1)
 dev.new()
 plot(Loading.x[,pcx], Loading.x[,pcy], col=col.pv, xlab = load.xlab, ylab = load.ylab, xlim = c(Min.pc1,Max.pc1), ylim = c(Min.pc2, Max.pc2), main = "PCA - Loading Plot (Significance-colored variables)", sub = "Variables in red showed Pvalue < 0.05")
 axis(1, at=Mpc1, pos=c(0,0), labels=FALSE, col="grey", lwd=0.7)
 axis(2, at=Mpc2, pos=c(0,0), labels=FALSE, col="grey", lwd=0.7)
 text(Loading.x[,pcx], Loading.x[,pcy], labels=rownames(Loading.x), cex=0.6, pos=1)
 E = paste(getwd(), "/PCA_Data_", scaling,"Loading_PC",pcx,"vsPC",pcy, "_Pvalues-colored.pdf", sep="" )
 dev.copy2pdf(file=E)
}