comparison edgeR_Differential_Gene_Expression.xml @ 3:12fb0d4b1e93 draft

planemo upload for repository https://bitbucket.org/EMCbioinf/galaxy-tool-shed-tools/raw/master/edger_with_design_matrix commit c7e4f2dfb8c35144b787850b60e116edfbaaa20f

2:ec951a5017f8

    <stdio>
        <regex match="Error in[^a-z]+contrasts"
               source="both"
               level="fatal"
               description="Have the design- and expression-matrix been swapped?" />
        <regex match="Execution halted"
               source="both"
               level="fatal" />
        <regex match="Calculating library sizes from column" 
               source="stderr" 

                $output_MDSplot_logFC
            #else:
                /dev/null
            #end if
            
            #if $output_MDSplot_bcv:
                $output_MDSplot_bcv
            #else:
                /dev/null
            #end if
            
            #if $output_BCVplot:

output_cpm                          <- args[6]

output_xpkm                         <- args[7]        ##FPKM file - to be implemented

output_raw_counts                   <- args[8]
output_MDSplot_logFC                <- args[9]
output_MDSplot_bcv                  <- args[10]
output_BCVplot                      <- args[11]
output_MAplot                       <- args[12]
output_PValue_distribution_plot     <- args[13]
output_hierarchical_clustering_plot <- args[14]
output_heatmap_plot                 <- args[15]
output_RData_obj                    <- args[16]
output_format_images                <- args[17]


## Obtain read-counts
expression_matrix <- read.delim(expression_matrix_file,header=T,stringsAsFactors=F,row.names=1,check.names=FALSE,na.strings=c(""))
design_matrix <- read.delim(design_matrix_file,header=T,stringsAsFactors=F,row.names=1,check.names=FALSE,na.strings=c(""))

  write("Estimating tagwise dispersion...",stdout())
  dge <- estimateGLMTagwiseDisp(dge,design)
  
  
  # hierarchical clustering makes use of the distance of the MDS
  if(output_MDSplot_logFC != "/dev/null" || output_hierarchical_clustering_plot != "/dev/null") {
    write("Calculating MDS plot (logFC method)",stdout())
    mds_distance_logFC <- plotMDS.DGEList(dge,top=500,labels=rep("",nrow(dge\$samples)))# Get coordinates of unflexible plot
    dev.off()# Kill it
    
    if(output_MDSplot_logFC != "/dev/null") {  
      write("Creating MDS plot (logFC method)",stdout())
      if(output_format_images == "pdf") {
        pdf(output_MDSplot_logFC,height=14,width=14)

      
      diff_x <- abs(max(mds_distance_logFC\$x)-min(mds_distance_logFC\$x))
      diff_y <-(max(mds_distance_logFC\$y)-min(mds_distance_logFC\$y))
      plot(c(min(mds_distance_logFC\$x),max(mds_distance_logFC\$x) + 0.45 * diff_x), c(min(mds_distance_logFC\$y) - 0.05 * diff_y,max(mds_distance_logFC\$y) + 0.05 * diff_y), main="edgeR logFC-MDS Plot on top 500 genes",type="n", xlab="Leading logFC dim 1", ylab="Leading logFC dim 2")
      points(mds_distance_logFC\$x,mds_distance_logFC\$y,pch=20)
      text(mds_distance_logFC\$x, mds_distance_logFC\$y,rownames(dge\$samples),cex=1.25,col="gray",pos=4)
      rm(diff_x,diff_y)
      
      dev.off()
      }
  }
  
  
  if(output_MDSplot_bcv != "/dev/null") {
    write("Creating MDS plot (bcv method)",stdout())
    
    ## 1. First create a virtual plot to obtain the desired coordinates
    pdf("bcvmds.pdf")
    mds_distance_BCV <- plotMDS.DGEList(dge,method="bcv",top=500,labels=rep("",nrow(dge\$samples)))
    dev.off()# Kill it
    
    ## 2. Re-plot the coordinates in a new figure with the size and settings.
    if(output_format_images == "pdf") {
      pdf(output_MDSplot_bcv,height=14,width=14)
    } else if(output_format_images == "svg") {
      svg(output_MDSplot_bcv,height=14,width=14)
    } else {
      ## png(output_MDSplot_bcv)
      ## png does not work out of the box in the Galaxy Toolshed Version of R due to its compile settings: https://biostar.usegalaxy.org/p/9170/
      
      bitmap(output_MDSplot_bcv,type="png16m",height=7*3,width=7*3)
    }
    
    diff_x <- abs(max(mds_distance_BCV\$x)-min(mds_distance_BCV\$x))
    diff_y <- (max(mds_distance_BCV\$y)-min(mds_distance_BCV\$y))
    plot(c(min(mds_distance_BCV\$x),max(mds_distance_BCV\$x) + 0.45 * diff_x), c(min(mds_distance_BCV\$y) - 0.05 * diff_y,max(mds_distance_BCV\$y) + 0.05 * diff_y), main="edgeR BCV-MDS Plot",type="n", xlab="Leading BCV dim 1", ylab="Leading BCV dim 2")
    points(mds_distance_BCV\$x,mds_distance_BCV\$y,pch=20)
    text(mds_distance_BCV\$x, mds_distance_BCV\$y,rownames(dge\$samples),cex=1.25,col="gray",pos=4)
    rm(diff_x,diff_y)
    
    dev.off()
  }
  
  
  if(output_BCVplot != "/dev/null") {
    write("Creating Biological coefficient of variation plot",stdout())

        <param name="fdr" type="float" min="0" max="1" value="0.05" label="False Discovery Rate (FDR)" />
        
        <param name="outputs" type="select" label="Optional desired outputs" multiple="true" display="checkboxes">
            <option value="make_output_raw_counts">Raw counts table</option>
            <option value="make_output_MDSplot_logFC">MDS-plot (logFC-method)</option>
            <option value="make_output_MDSplot_bcv">MDS-plot (BCV-method; much slower)</option>
            <option value="make_output_BCVplot">BCV-plot</option>
            <option value="make_output_MAplot">MA-plot</option>
            <option value="make_output_PValue_distribution_plot">P-Value distribution plot</option>
            <option value="make_output_hierarchical_clustering_plot">Hierarchical custering</option>
            <option value="make_output_heatmap_plot">Heatmap</option>

                <when input="output_format_images" value="pdf" format="pdf" />
                <when input="output_format_images" value="svg" format="svg" />
            </change_format>
        </data>
        
        <data format="png" name="output_MDSplot_bcv" label="edgeR DGE on ${design_matrix.hid}: ${design_matrix.name} - MDS-plot (bcv method)">
            <filter>outputs and ("make_output_MDSplot_bcv" in outputs)</filter>
            
            <change_format>
                <when input="output_format_images" value="png" format="png" />
                <when input="output_format_images" value="pdf" format="pdf" />
                <when input="output_format_images" value="svg" format="svg" />
            </change_format>
        </data>
        
        <data format="png" name="output_BCVplot" label="edgeR DGE on ${design_matrix.hid}: ${design_matrix.name} - BCV-plot">
            <filter>outputs and ("make_output_BCVplot" in outputs)</filter>
            

3:12fb0d4b1e93

    <stdio>
        <regex match="Error in[^a-z]+contrasts"
               source="both"
               level="fatal"
               description="Have the design- and expression-matrix been swapped?" />
        <regex match="Error in eval\(expr, envir, enclos\)"
               source="both"
               level="fatal"
               description="You have most likely used a condition in the contrast, that is not present in the Design matrix" />
        <regex match="Execution halted"
               source="both"
               level="fatal" />
        <regex match="Calculating library sizes from column" 
               source="stderr" 

                $output_MDSplot_logFC
            #else:
                /dev/null
            #end if
            
            #if $output_MDSplot_logFC_coordinates:
                $output_MDSplot_logFC_coordinates
            #else:
                /dev/null
            #end if
            
            #if $output_MDSplot_bcv:
                $output_MDSplot_bcv
            #else:
                /dev/null
            #end if
            
            #if $output_MDSplot_bcv_coordinates:
                $output_MDSplot_bcv_coordinates
            #else:
                /dev/null
            #end if
            
            #if $output_BCVplot:

output_cpm                          <- args[6]

output_xpkm                         <- args[7]        ##FPKM file - to be implemented

output_raw_counts                   <- args[8]

output_MDSplot_logFC                <- args[9]
output_MDSplot_logFC_coordinates    <- args[10]

output_MDSplot_bcv                  <- args[11]
output_MDSplot_bcv_coordinates      <- args[12]

output_BCVplot                      <- args[13]
output_MAplot                       <- args[14]
output_PValue_distribution_plot     <- args[15]
output_hierarchical_clustering_plot <- args[16]
output_heatmap_plot                 <- args[17]
output_RData_obj                    <- args[18]
output_format_images                <- args[19]


## Obtain read-counts
expression_matrix <- read.delim(expression_matrix_file,header=T,stringsAsFactors=F,row.names=1,check.names=FALSE,na.strings=c(""))
design_matrix <- read.delim(design_matrix_file,header=T,stringsAsFactors=F,row.names=1,check.names=FALSE,na.strings=c(""))

  write("Estimating tagwise dispersion...",stdout())
  dge <- estimateGLMTagwiseDisp(dge,design)
  
  
  # hierarchical clustering makes use of the distance of the MDS
  if(output_MDSplot_logFC != "/dev/null" || output_MDSplot_logFC_coordinates != "/dev/null" || output_hierarchical_clustering_plot != "/dev/null") {
    write("Calculating MDS plot (logFC method)",stdout())
    n_dim <- nrow(dge\$samples)
    mds_distance_logFC <- plotMDS.DGEList(dge,top=500,labels=rep("",nrow(dge\$samples)),dim.plot=c(n_dim-2,n_dim-1))
    dev.off()# Kill it
    
    # Reset to primary dimensions
    mds_distance_logFC\$x = mds_distance_logFC\$cmdscale.out[,1]
    mds_distance_logFC\$y = mds_distance_logFC\$cmdscale.out[,2]
    
    if(output_MDSplot_logFC != "/dev/null") {  
      write("Creating MDS plot (logFC method)",stdout())
      if(output_format_images == "pdf") {
        pdf(output_MDSplot_logFC,height=14,width=14)

      
      diff_x <- abs(max(mds_distance_logFC\$x)-min(mds_distance_logFC\$x))
      diff_y <-(max(mds_distance_logFC\$y)-min(mds_distance_logFC\$y))
      plot(c(min(mds_distance_logFC\$x),max(mds_distance_logFC\$x) + 0.45 * diff_x), c(min(mds_distance_logFC\$y) - 0.05 * diff_y,max(mds_distance_logFC\$y) + 0.05 * diff_y), main="edgeR logFC-MDS Plot on top 500 genes",type="n", xlab="Leading logFC dim 1", ylab="Leading logFC dim 2")
      points(mds_distance_logFC\$x,mds_distance_logFC\$y,pch=20)
      text(mds_distance_logFC\$x,mds_distance_logFC\$y,rownames(dge\$samples),cex=1.25,col="gray",pos=4)
      rm(diff_x,diff_y)
      
      dev.off()
    }
    
    if(output_MDSplot_logFC_coordinates != "/dev/null") {
      n_dim <- ncol(mds_distance_logFC\$cmdscale.out)
      colnames(mds_distance_logFC\$cmdscale.out) <- paste(rep("Dim",n_dim),(1:n_dim),sep="_")
      export <- data.frame(samples=rownames(mds_distance_logFC\$cmdscale.out),mds_distance_logFC\$cmdscale.out)
      row.names(export) <- NULL
      write.table(file=output_MDSplot_logFC_coordinates,export,sep="\t",row.names=FALSE,col.names=TRUE)
    }
  }
  
  
  if(output_MDSplot_bcv != "/dev/null" || output_MDSplot_bcv_coordinates != "/dev/null" ) {
    write("Creating MDS plot (bcv method)",stdout())
    
    ## 1. First create a virtual plot to obtain the desired coordinates
    n_dim <- nrow(dge\$samples)
    mds_distance_BCV <- plotMDS.DGEList(dge,method="bcv",top=500,labels=rep("",nrow(dge\$samples)),dim.plot=c(n_dim-2,n_dim-1))
    dev.off()
    
    if(output_MDSplot_logFC != "/dev/null") {  
      ## 2. Re-plot the coordinates in a new figure with the size and settings.
      if(output_format_images == "pdf") {
        pdf(output_MDSplot_bcv,height=14,width=14)
      } else if(output_format_images == "svg") {
        svg(output_MDSplot_bcv,height=14,width=14)
      } else {
        ## png(output_MDSplot_bcv)
        ## png does not work out of the box in the Galaxy Toolshed Version of R due to its compile settings: https://biostar.usegalaxy.org/p/9170/
        
        bitmap(output_MDSplot_bcv,type="png16m",height=7*3,width=7*3)
      }
      
      diff_x <- abs(max(mds_distance_BCV\$x)-min(mds_distance_BCV\$x))
      diff_y <- (max(mds_distance_BCV\$y)-min(mds_distance_BCV\$y))
      plot(c(min(mds_distance_BCV\$x),max(mds_distance_BCV\$x) + 0.45 * diff_x), c(min(mds_distance_BCV\$y) - 0.05 * diff_y,max(mds_distance_BCV\$y) + 0.05 * diff_y), main="edgeR BCV-MDS Plot",type="n", xlab="Leading BCV dim 1", ylab="Leading BCV dim 2")
      points(mds_distance_BCV\$x,mds_distance_BCV\$y,pch=20)
      text(mds_distance_BCV\$x, mds_distance_BCV\$y,rownames(dge\$samples),cex=1.25,col="gray",pos=4)
      rm(diff_x,diff_y)
      
      dev.off()
    }
    
    if(output_MDSplot_bcv_coordinates != "/dev/null") { 
      n_dim <- ncol(mds_distance_BCV\$cmdscale.out)
      colnames(mds_distance_BCV\$cmdscale.out) <- paste(rep("Dim",n_dim),(1:n_dim),sep="_")
      export <- data.frame(samples=rownames(mds_distance_BCV\$cmdscale.out),mds_distance_BCV\$cmdscale.out)
      row.names(export) <- NULL
      write.table(file=output_MDSplot_bcv_coordinates,export,sep="\t",row.names=FALSE,col.names=TRUE)
    }
  }
  
  
  if(output_BCVplot != "/dev/null") {
    write("Creating Biological coefficient of variation plot",stdout())

        <param name="fdr" type="float" min="0" max="1" value="0.05" label="False Discovery Rate (FDR)" />
        
        <param name="outputs" type="select" label="Optional desired outputs" multiple="true" display="checkboxes">
            <option value="make_output_raw_counts">Raw counts table</option>
            <option value="make_output_MDSplot_logFC">MDS-plot (logFC-method)</option>
            <option value="make_output_MDSplot_logFC_coordinates">MDS-plot coordinates table (logFC-method)</option>
            <option value="make_output_MDSplot_bcv">MDS-plot (BCV-method; slow)</option>
            <option value="make_output_MDSplot_bcv_coordinates">MDS-plot coordinates table (BCV-method; slow)</option>
            <option value="make_output_BCVplot">BCV-plot</option>
            <option value="make_output_MAplot">MA-plot</option>
            <option value="make_output_PValue_distribution_plot">P-Value distribution plot</option>
            <option value="make_output_hierarchical_clustering_plot">Hierarchical custering</option>
            <option value="make_output_heatmap_plot">Heatmap</option>

                <when input="output_format_images" value="pdf" format="pdf" />
                <when input="output_format_images" value="svg" format="svg" />
            </change_format>
        </data>
        
        <data format="tabular" name="output_MDSplot_logFC_coordinates" label="edgeR DGE on ${design_matrix.hid}: ${design_matrix.name} - MDS-plot coordinates table (logFC method)">
            <filter>outputs and ("make_output_MDSplot_logFC_coordinates" in outputs)</filter>
        </data>
        
        <data format="png" name="output_MDSplot_bcv" label="edgeR DGE on ${design_matrix.hid}: ${design_matrix.name} - MDS-plot (bcv method)">
            <filter>outputs and ("make_output_MDSplot_bcv" in outputs)</filter>
            
            <change_format>
                <when input="output_format_images" value="png" format="png" />
                <when input="output_format_images" value="pdf" format="pdf" />
                <when input="output_format_images" value="svg" format="svg" />
            </change_format>
        </data>
        
        <data format="tabular" name="output_MDSplot_bcv_coordinates" label="edgeR DGE on ${design_matrix.hid}: ${design_matrix.name} - MDS-plot coordinates table (BCV method)">
            <filter>outputs and ("make_output_MDSplot_bcv_coordinates" in outputs)</filter>
        </data>
        
        <data format="png" name="output_BCVplot" label="edgeR DGE on ${design_matrix.hid}: ${design_matrix.name} - BCV-plot">
            <filter>outputs and ("make_output_BCVplot" in outputs)</filter>