Mercurial > repos > blankenberglab > gpca

#!/usr/bin/env Rscript
library(optparse)
library(gPCA)
options(bitmapType='cairo')


# parse options
option_list = list(
   make_option(
    c("--version"),
    action = "store_true",
    default = FALSE,
    help = "Print version and exit"
   ),
  make_option(
    c("-i", "--batch"),
    action = "store",
    default = NA,
    type = 'character',
    help = "Input batch"
  ),
  make_option(
    c("-d", "--data"),
    action = "store",
    default = NA,
    type = 'character',
    help = "Input data"
  ),
   make_option(
    c("-n", "--nperm"),
    action = "store",
    default = 1000,
    type = 'numeric',
    help = "Number of permutations to preform"
   ),
   make_option(
    c("-c", "--center"),
    action = "store_true",
    default = FALSE,
    help = "Is the input batch and data centered [default %default]"
	 ),
   make_option(
    c("-y", "--scaleY"),
    action = "store_true",
    default = FALSE,
    help = "Scale to Y axis"
   ),
   make_option(
    c("-f", "--filt"),
    action = "store",
    default = NULL,
    type = 'numeric',
    help = "Retain features"
	 ),
   make_option(
    c( "--npcs"),
    action = "store",
    default = 3,
    type = 'numeric',
    help = "Number of principal components to plot"
   ),
   make_option(
    c("-p", "--path"),
    action = "store",
    default = '$html_outfile.extra_files_path',
    type = 'character',
    help = "File path"
   ),
   make_option(
    c("-s", "--seed"),
    action = "store",
    default = NULL,
    type = 'numeric',
    help = "Set a seed number"
   ),
   make_option(
    c("-x", "--numbers_outfile"),
    action = "store",
    default = NA,
    type = 'character',
    help = "Numbers output"
 	 ),
   make_option(
    c("-o", "--html_outfile"),
    action = "store",
    default = NA,
    type = 'character',
    help = "Output",
  )
)

opt <-parse_args(OptionParser(option_list = option_list))

if (opt$version){
  # print version and exit
  cat(paste("gPCA version", toString(packageVersion("gPCA"))), "\n")
  quit()
}

# Check parameter values

if ( ! file.exists(opt$batch)){
  stop((paste('File for batch', opt$batch, 'does not exist')))
}
if ( ! file.exists(opt$data)){
  stop((paste('File for data', opt$data, 'does not exist')))
}

dir.create(opt$path);
data<-t(as.matrix(read.delim(opt$data, row.names=1, header=TRUE, sep="\t")))
batch<-as.matrix(read.delim(opt$batch, header=TRUE, row.names=1, sep="\t"))

write('#Batch\tDelta\tP-value\tCumulative Variance',file = opt$numbers_outfile, append=TRUE)

write('<html><body><div align="left" id="top"><p><h1>Table of Contents</h1><ul>',file = opt$html_outfile, append=TRUE)
for (row in 1:nrow(batch)) {
  write(paste0('<li><a href="#',row.names(batch)[row],'">',row.names(batch)[row],'</a></li>'), file=opt$html_outfile, append=TRUE)
}
write('</ul></p></div>', file=opt$html_outfile, append=TRUE)

for (row in 1:nrow(batch)) {
  batch1<-as.numeric(factor(batch[row,]))

out<-gPCA.batchdetect(x=data, batch=batch1, center=opt$center, scaleY=opt$scaleY, nperm=opt$nperm, filt=opt$filt, seed=opt$seed)

out$varPCg1<-((out$varPCg1-out$varPCu1)/out$varPCg1)*100

write(paste(row.names(batch)[row],out$delta,out$p.val,out$varPCg1,sep="\t"),file = opt$numbers_outfile, append=TRUE)

# General Distribution
par(mai=c(0.8,0.8,0.1,0.1),cex=0.8)
png(paste(opt$path,'/gDist_',row,'.png', sep=""), width=1020, height=800, units='px')
gDist(out)
invisible(dev.off())


# Guided/Unguided 1v2
par(mai=c(0.8,0.8,0.1,0.1),cex=0.8)
png(paste(opt$path,'/guided_1v2_',row,'.png', sep=""), width=1020, height=800, units='px')
PCplot(out,ug="unguided",type="1v2")
invisible(dev.off())

par(mai=c(0.8,0.8,0.1,0.1),cex=0.8)
png(paste(opt$path,'/unguided_1v2_',row,'.png', sep=""), width=1020, height=800, units='px')
PCplot(out,ug="unguided",type="1v2")
invisible(dev.off())


# Guided/Unguided comp,3
par(mai=c(0.65,0.65,0.1,0.1),cex=0.8)
png(paste(opt$path,'/guided_npcs_',row,'.png', sep=""), width=1020, height=800, units='px')
PCplot(out,ug="guided",type="comp",npcs=opt$npcs)
invisible(dev.off())

par(mai=c(0.65,0.65,0.1,0.1),cex=0.8)
png(paste(opt$path,'/unguided_npcs_',row,'.png', sep=""), width=1020, height=800, units='px')
PCplot(out,ug="unguided",type="comp",npcs=opt$npcs)
invisible(dev.off())


# Guided/Unguided CumlativeVarPlot
par(mai=c(0.8,0.8,0.1,0.1),cex=0.8)
png(paste(opt$path,'/guided_var_',row,'.png', sep=""), width=1020, height=800, units='px')
CumulativeVarPlot(out,ug="guided",col="blue")
invisible(dev.off())

par(mai=c(0.8,0.8,0.1,0.1),cex=0.8)
png(paste(opt$path,'/unguided_var_',row,'.png', sep=""), width=1020, height=800, units='px')
CumulativeVarPlot(out,ug="unguided",col="blue")
invisible(dev.off())


write(paste0('
<header id=',row.names(batch)[row],'><center><font size="11">',row.names(batch)[row],'</font><center></header>
<br>
<table align="center" border="1">
 <tr>
  <th>Delta</th><th>P-value</th><th>varPCg1</th>
 </tr>
 <tr>
  <td id="#">',out$delta,'</td><td id=p.val>',out$p.val,'</td><td id=varPCg1>',out$varPCg1,'</td>
 </tr>
</table>
<center><img src="gDist_',row,'.png"/></center><br>
<center><title>Guided 1v2</title><br><img src="guided_1v2_',row,'.png"/></center><br>
<center><title>Unguided 1v2</title><br><img src="unguided_1v2_',row,'.png"/></center><br>

<center><title>Guided Compare to ',opt$npcs,'</title><br><img src="guided_npcs_',row,'.png"/></center><br>
<center><title>Unguided Compare to ',opt$npcs,'</title><br><img src="unguided_npcs_',row,'.png"/></center><br>

<center><title>Guided Cumulative Variance</title><br><img src="guided_var_',row,'.png"/></center><br>
<center><title>Unguided Cumulative Variance</title><br><img src="unguided_var_',row,'.png"/></center><br>

'), file = opt$html_outfile, append=TRUE)
}
write(paste('<p><center><a href="#top">Back to Top</a></center></p></body></html>'),file = opt$html_outfile, append=TRUE)

sessionInfo()
author	blankenberglab
date	Tue, 27 Nov 2018 12:29:58 -0500
parents	d93629e79f5e
children	8def7dda3bee