PCA

#Here the object data is the gene expression from RMA with dimension pxn = 54675 x 248, here n=248 (two batches: 192+56)

genes<-data[1:54613,]

genes<-t(genes)

pcs<-prcomp(genes)

summary(pcs)

library(scatterplot3d)

PC1<-pcs$x[,1]

PC2<-pcs$x[,2]

PC3<-pcs$x[,3]

group.colors <- rep("#000000", length(PC1))

group.colors[seq(1,192,1)] = "#FF77FF"   #SG Batch A

group.colors[seq(193,248,1)] = "blue"      #SG Batch B

scatterplot3d(PC3,PC1,PC2, main="PCA scatterplot before ComBat normalization", color=group.colors, pch=16)

legend.txt <- c("SG Batch A", "SG Batch B")

legend.col <- c("#00FF00","#0000FF")

legend(-5.75, 5.2, legend.txt , bty="n", col=legend.col,cex=1.2, pch=15)