view PCA_plot.R @ 16:fea55f4efcce draft

Uploaded
author mb2013
date Tue, 20 May 2014 03:29:11 -0400
parents 2e8c9032e8d8
children
line wrap: on
line source

#Plottool makes a graph of Principal Components created with a Principal Component analysis. 
#MB

#commands extracting of commandline
args <- commandArgs(TRUE)

#input files and options
input <- args[1]
main_title <- args[2]
x_title <- args[3]
y_title <- args[4]
x_column <- args[5]
y_column <- args[6]
names <- args [7] #name of every sample in one file
#output file
output <- args[8]

suppressMessages(library("geomorph")) #package geomorph

#reading of input files
read <- read.csv(file <- input,header = TRUE)
read2 <- scan(file <- names, what = "", quiet = TRUE)

pca1 <- read[,as.integer(x_column)] #principal component
pca2 <- read[,as.integer(y_column)] #principal component

png(output) #output in png format

#axis boundaries
minpca1 = min(pca1) - max(pca1)
maxpca1 = max(pca1) + max(pca1)
minpca2 = min(pca2) - max(pca2)
maxpca2 = max(pca2) + max(pca2)

#creating the plot with principal components and titels
suppressMessages(plot(pca1,pca2, main = main_title, xlab = x_title, ylab = y_title, pch=20,cex=0.6, xlim = c(minpca1,maxpca1), ylim=c(minpca2,maxpca2)))
#add labels to data points
text(pca1,pca2,labels = read2, pos = 2, cex = 0.7,col = heat.colors(35:40))

graphics.off()