view cluster.tools/normalize.matrix.R @ 9:a3c03541fe6f draft default tip

Uploaded

#!/usr/bin/env Rscript
##  Script by Peter Waltman
## 
## License under Creative Commons Attribution 3.0 Unported (CC BY 3.0)
##
argspec <- c("TBD


        Usage: 
                normalize.matrix.R -d <data.file> 
        Optional:
                -o <output.name>
                TBD 
                \n\n")

args <- commandArgs(TRUE)
if ( length( args ) == 1 && args =="--help") { 
  write(argspec, stderr())
  q();
}

## some helper fn's
write.2.tab <- function( mat,
                         fname ) {
  mat <- rbind( colnames( mat ), mat )
  mat <- cbind( c( "ID", rownames( mat )[-1] ),
                      mat )
  write.table( mat, fname, sep="\t", row.names=FALSE, col.names=FALSE, quote=FALSE )
}

lib.load.quiet <- function( package ) {
   package <- as.character(substitute(package))
   suppressPackageStartupMessages( do.call( "library", list( package=package ) ) )
}
lib.load.quiet(getopt)


spec <- matrix( c( "data.fname",         "d", 1, "character",
                   "center.rows",        "r", 2, "character",
                   "var.adj.rows",       "R", 2, "character",
                   "center.cols",        "c", 2, "character",
                   "var.adj.cols",       "C", 2, "character",
                   "output.fname",       "o", 2, "character"
                   ),
                nc=4,
                byrow=TRUE
               )

opt <- getopt( spec=spec )

#set some reasonable defaults for the options that are needed,
#but were not specified.
if ( is.null( opt$output.fname ) ) {
  out.fname <- ""
  if ( ! is.null( opt$center.rows ) ) out.fname <- paste( "row", opt$center.rows, "centered", out.fname, sep="." )
  if ( ! is.null( opt$center.cols ) ) out.fname <- paste( "col", opt$center.cols, "centered", out.fname, sep="." )
  if ( ! is.null( opt$var.adj.rows ) ) out.fname <- paste( "row", opt$var.adj.rows, "var.adjed", out.fname, sep="." )
  if ( ! is.null( opt$var.adj.cols ) ) out.fname <- paste( "col", opt$var.adj.cols, "var.adjed", out.fname, sep="." )
  
  opt$output.fname <- paste( out.fname, "centered.matrix", sep="." )
}
if ( is.null( opt$center.rows ) ) { opt$center.rows <- 'none' }
if ( is.null( opt$center.cols ) ) { opt$center.cols <- 'none' }
if ( is.null( opt$var.adj.rows ) ) { opt$var.adj.rows <- 'none' }
if ( is.null( opt$var.adj.cols ) ) { opt$var.adj.cols <- 'none' }

data <- as.matrix( read.delim( opt$data.fname, header=T, row.names=1 , check.names=FALSE ) )
my.center <- rep( 0, nrow( data ) )
my.var.adj <- rep( 1, nrow( data ) )
if ( opt$center.rows != "none" ) {
  my.center.fn <- get( opt$center.rows )
  ##data <- sweep( data, 1, apply( data, 1, my.center.fn, na.rm=T ) )
  my.center <- apply( data, 1, my.center.fn, na.rm=T )
}
if ( opt$var.adj.rows != "none" ) {
  my.var.adj.fn <- get( opt$var.adj.rows )
  my.var.adj <- apply( data, 1, my.var.adj.fn, na.rm=T )
}
data <- t( scale( t( data ), center=my.center, scale=my.var.adj ) )

my.center <- rep( 0, ncol( data ) )
my.var.adj <- rep( 1, ncol( data ) )
if ( opt$center.cols != "none" ) {
  my.center.fn <- get( opt$center.cols )
  ##data <- sweep( data, 2, apply( data, 2, my.center.fn, na.rm=T ) )
  my.center <- apply( data, 2, my.center.fn, na.rm=T )
}
my.var.adj <- rep( 1, ncol( data ) )
if ( opt$var.adj.cols != "none" ) {
  my.var.adj.fn <- get( opt$var.adj.cols )
  my.var.adj <- apply( data, 2, my.var.adj.fn, na.rm=T )
}
data <- scale( data, center=my.center, scale=my.var.adj )

write.2.tab( data, opt$output.fname )