comparison cluster.tools/rdata.2.out.R @ 0:0decf3fd54bc draft

Uploaded

0:0decf3fd54bc

#!/usr/bin/env Rscript
argspec <- c("tab.2.cdt.R converts a data matrix to cdt format

        Usage: 
                tab.2.cdt.R -d <data.file> 
        Optional:
                            -o <output_file>
                \n\n")
args <- commandArgs(TRUE)
if ( length( args ) == 1 && args =="--help") { 
  write(argspec, stderr())
  q();
}

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

lib.load.quiet( getopt )
lib.load.quiet( ctc )
if ( any( c( 'flashClust', 'fastcluster' ) %in% installed.packages() ) ) {
  if ( 'flashClust' %in% installed.packages() ) {
    lib.load.quiet( flashClust )
  } else {
    if ( 'fastcluster' %in% installed.packages() ) {
      lib.load.quiet( fastcluster )
    }
  }
}

spec <- matrix( c( "dataset",             "d", 1, "character",
                   "dataset2",            "D", 2, "character",
                   "output.format",       "f", 2, "character",
                   "output.report.dir",   "p", 2, "character",
                   "output.fname",        "o", 2, "character"
                   ),
                nc=4,
                byrow=TRUE
               )


opt <- getopt( spec=spec )
if ( is.null( opt$output.report.dir ) ) { opt$output.report.dir <- "report" }
if ( is.null( opt$output.fname ) ) { opt$output.fname <- file.path( opt$output.report.dir, paste( "data", opt$output.format, sep="." ) ) }
if ( is.null( opt$output.format ) ) { opt$output.format <- "cdt" }


load( opt$dataset )  ## should load the cl, treecl.res (or partcl.res) and data

if ( opt$output.format %in% c( "cls-only", "newick" ) ) {
  if ( opt$output.format == "cls-only" ) {

    cl <- cbind( names( cl ), as.numeric( cl ) )
    colnames( cl ) <- c( "ID", "Class" )

    opt$output.fname <- gsub( "cls-only$", "tab", opt$output.fname )
    write.table( cl, opt$output.fname, sep="\t", quote=FALSE, row.names=FALSE, col.names=FALSE )
  } else {
    ##if ( opt$output.format == "newick" ) {

    if ( ! exists( "treecl.res" ) ) stop( "no HAC result found in results file proved - necessary to generate a Newick formated file.\n" )
    write( hc2Newick( treecl.res ), opt$output.fname )
  }
} else {
  if ( ! exists( 'data' ) ) stop( "No data object in the rdata file provided for", opt$output.format, "format!!\n" )
  if ( inherits( data, "dist" ) ) stop( "data provided is a distance matrix - not a data matrix.  Can't generate TreeView or Tab-delimited files w/distance matrices!\n" )

  ## the rest of this is for the remaining output formats
  ##  pre-set the cluster results for rows & cols to NULL
  hr <- hr.cl <- hc <- hc.cl <- NULL
  if ( exists( 'treecl.res' ) ) {

    if ( is.null( treecl.res$dist.method ) ) treecl.res$dist.method <- 'euclidean'  # just set it to some stub so that the ctc fn's don't complain
    if ( all( names( cl ) %in% rownames( data ) ) ) {
      hr <- treecl.res
      hr.cl <- cl
    } else if ( all( names( cl ) %in% colnames( data ) ) ) {
      hc <- treecl.res
      hc.cl <- cl
    } else {
      stop( "Specified cluster result does not come from this data set\n" )
    }

  } else {
    if ( exists( 'partcl.res' ) ) {
      if ( all( names( cl ) %in% rownames( data ) ) ) {
        hr <- NA
        hr.cl <- cl
        orig.data <- data
        data <- data[ names( cl ), ]  ## partcl.res should now be sorted in order of cluster
      } else if ( all( names( cl ) %in% colnames( data ) ) ) {
        hc <- NA
        hc.cl <- cl
        orig.data <- data
        data <- data[ , names( cl ) ]  ## partcl.res should now be sorted in order of cluster
      } else {
        stop( "Specified cluster result does not come from this data set\n" )
      }
    }
    else {
      stop( 'could not find a valid cluster result to use for primary direction\n' )
    }
  }
  

  if ( ! is.null( opt$dataset2 ) ) {
    
    ## prep for loading new cluster result
    if ( ! exists( 'orig.data' ) ) orig.data <- data
    if ( exists( "treecl.res" ) ) {
      rm( treecl.res )
    } else if ( exists( "partcl.res" ) ) {
      rm( partcl.res )
    } else stop( "no primary clustering found when generating the 2nd\n" )
    rm( cl, data )
    
  
    load( opt$dataset2 ) ## this should bring in the cl obj for the 2nd direction

    ## check the data 1st
    if ( length( orig.data ) != length( data ) ) stop( "incompatible cluster results in 2nd results file - matrices are diff lengths\n" )
    if ( nrow( orig.data ) != nrow( data ) ) stop( "incompatible cluster results in 2nd results file - matrices have diff dimensions\n" )
    if ( sum( orig.data == data ) != length( orig.data ) )  stop( "incompatible cluster results in 2nd results file - matrices contain diff contents\n" )
    ## looks like data is the same, so drop a copy and start chugging
    rm( orig.data ); gc()
    
    if ( exists( 'treecl.res' ) ) {
      if ( is.null( treecl.res$dist.method ) ) treecl.res$dist.method <- 'euclidean'  # just set it to some stub so that the ctc fn's don't complain
      
      if ( is.null( hr ) ) {
        if ( all( rownames( cl ) %in% rownames( data ) ) ) {
          hr <- treecl.res
          hr.cl <- cl
        } else {
          stop( "results file for 2nd direction doesn't contain cluster for 2ndary direction (rows in this case)\n" )
        }
      } else if ( is.null( hc ) ) {
        if ( all( rownames( cl ) %in% colnames( data ) ) ) {
          hc <- treecl.res
          hc.cl <- cl
        } else {
          stop( "results file for 2nd direction doesn't contain cluster for 2ndary direction (genes in this case)\n" )
        }
      } else {
        stop( "should never get here\n" )
      }
    } else if ( exists( 'partcl.res' ) ) {
      if ( is.null( hr ) ) {
        if ( all( names( cl ) %in% rownames( data ) ) ) {
          hr <- NA
          hr.cl <- cl
          data <- data[ names( cl ), ]  ## partcl.res should now be sorted in order of cluster
        } else {
          stop( "results file for 2nd direction doesn't contain cluster for 2ndary direction (rows in this case)\n" )
        }
      } else if ( is.null( hc ) ) {
        if ( all( names( cl ) %in% colnames( data ) ) ) {
          hc <- NA
          hc.cl <- cl
          data <- data[ , names( cl ) ]  ## partcl.res should now be sorted in order of cluster
        } else {
          stop( "results file for 2nd direction doesn't contain cluster for 2ndary direction (genes in this case)\n" )
        }
      } else {
        stop( "should never get here\n" )
      }      
    }
  }
  
  ## Now, re-set hc & nr to NULL if they were set to NA
  ## we used NA to signify that they were set by kmeans/pam, but now, we need to reset them
  ## for the following lines (that generate the dendrograms (if there was an hclust result)
  if ( ( !is.null( hr ) ) && is.na( hr ) ) hr <- NULL
  if ( ( !is.null( hc ) ) && is.na( hc ) ) hc <- NULL
  
  if ( ! exists( 'data' ) ) stop( "No data object in the rdata file provided!!\n" )
  
  if ( is.null( hc ) ) hc <- list( order=1:ncol( data ) )
  if ( is.null( hr ) ) hr <- list( order=1:nrow( data ) )

  if ( opt$output.format == "tabular" ) {
    write.table( data[ hr$order, hc$order ], opt$output.fname, quote=FALSE, sep="\t", col.names=NA )
  } else if ( opt$output.format == "cdt" ) {
    if ( !file.exists( opt$output.report.dir ) ){
      dir.create(opt$output.report.dir, recursive=T)
    }
    
    treeview.fname.stem <- file.path( opt$output.report.dir, "cluster.heatmap")
    fnames <- character()
    if ( inherits( hr, "hclust" ) ) {
      fname <- paste( treeview.fname.stem, ".gtr", sep="" )
      ## we manually specify a 'stub' distance b/c o/w it'll try using the attr(hr,"method")
      ##  and the r2gtr fn's get grumpy if the distance was anything starting with a 'p'
      r2gtr( hr, file=fname, distance='stub' )  
      fnames <- c( fnames, fname )
    }
    if ( inherits( hc, "hclust" ) ) {
      fname <- paste( treeview.fname.stem, ".atr", sep="" )
      r2atr( hc, file=fname, distance='stub' )
      fnames <- c( fnames, fname )
    }

    fname <- paste( treeview.fname.stem, ".cdt", sep="" )
    r2cdt( hr, hc, data, file=fname )
    fnames <- c( fnames, fname )

    ## jtv file now
    jtv.str <- '<DocumentConfig><UrlExtractor/><ArrayUrlExtractor/><Views><View type="Dendrogram" dock="1"><ColorExtractor contrast="2.0"><ColorSet zero="#FFFFFF" down="#0000FF"/></ColorExtractor><ArrayDrawer/><GlobalXMap current="Fill"><FixedMap type="Fixed"/><FillMap type="Fill"/><NullMap type="Null"/></GlobalXMap><GlobalYMap current="Fill"><FixedMap type="Fixed"/><FillMap type="Fill"/><NullMap type="Null"/></GlobalYMap><ZoomXMap><FixedMap type="Fixed"/><FillMap type="Fill"/><NullMap type="Null"/></ZoomXMap><ZoomYMap><FixedMap type="Fixed"/><FillMap type="Fill"/><NullMap type="Null"/></ZoomYMap><TextView><TextView face="Monospaced" size="14"><GeneSummary/></TextView><TextView face="Monospaced" size="14"><GeneSummary/></TextView><TextView face="Monospaced" size="14"><GeneSummary/></TextView><TextView face="Monospaced" size="14"><GeneSummary/></TextView></TextView><ArrayNameView face="Monospaced" size="14"><ArraySummary included="0"/></ArrayNameView><AtrSummary/><GtrSummary/></View></Views></DocumentConfig>'
    fname <- paste( treeview.fname.stem, ".jtv", sep="" )
    cat( jtv.str, file=fname )
    fnames <- c( fnames, fname )

    cmd <- paste( "tar -zcf",
                  opt$output.fname,
                  paste( "--directory=", opt$output.report.dir, sep="" ),
                  paste( basename( fnames ), collapse=" " ) )
    system( cmd )
  }
}