comparison combine.xml @ 1:65245dc812c3 draft

planemo upload for repository https://github.com/galaxyproteomics/tools-galaxyp/tree/master/tools/cardinal commit d2f311f7fff24e54c565127c40414de708e31b3c

0:93be1d20b5c3

<tool id="cardinal_combine" name="MSI combine" version="@VERSION@.0">
    <description>
        combine several mass spectrometry imaging datasets into one
    </description>
    <macros>
        <import>macros.xml</import>
    </macros>
    <expand macro="requirements">
        <requirement type="package" version="3.0.0">r-ggplot2</requirement>
    </expand>
    <command detect_errors="exit_code">
    <![CDATA[
        #for $i, $infile in enumerate($infiles):
            #if $infile.ext == 'imzml'

                ln -s '$annotation_file' annotation_file_${i}.tabular &&
            #end for
        #end if

        cat '${msi_combine}' &&
        Rscript '${msi_combine}'

    ]]>
    </command>
    <configfiles>
        <configfile name="msi_combine"><![CDATA[

    ## read and manipulate MSI data

    #if $infile.ext == 'imzml'
        #if str($processed_cond.processed_file) == "processed":
            msidata_$i <- readImzML('infile_${i}', mass.accuracy=$processed_cond.accuracy, units.accuracy = "$processed_cond.units")
            centroided(msidata) = $centroids
        #else
            msidata_$i <- readImzML('infile_${i}')
            centroided(msidata) = $centroids
        #end if
    #elif $infile.ext == 'analyze75'
            msidata_$i <- readAnalyze('infile_${i}')
            centroided(msidata) = $centroids
    #else
            msidata_$i = loadRData('infile_${i}.RData')
    #end if

    ## remove duplicated coordinates, otherwise combine will fail

        msidata_$i\$column3 = rep(NA, ncol(msidata_$i))
        msidata_$i\$column4 = rep(NA, ncol(msidata_$i))
        msidata_$i\$column5 = rep(NA, ncol(msidata_$i))
        msidata_$i\$combined_sample = rep(NA, ncol(msidata_$i))

    #if str($annotation_cond.annotation_tabular) == 'annotation'
        print("annotations")

        ## read annotation tabular, set first two columns as x and y, merge with coordinates dataframe and order according to pixelorder in msidata
        input_annotation = read.delim("annotation_file_${i}.tabular", header = $annotation_cond.tabular_header, stringsAsFactors = FALSE)

        msidata_$i\$column2 = annotation_df_8_sorted[,5]
        msidata_$i\$column3 = annotation_df_8_sorted[,6]
        msidata_$i\$column4 = annotation_df_8_sorted[,7]
        msidata_$i\$column5 = annotation_df_8_sorted[,8]

        ## extract columnnames from (last) annotation tabular (for QC plot names)
        annotation_colnames = colnames(input_annotation)[-c(1,2)]

    #end if

    ################### preparation xy shifts ##########################

    #if str( $combine_conditional.combine_method ) == 'xy_shifts':

        ## shift coordinates according to input tabular file and store file names
        coord(msidata_$i)\$x = coord(msidata_$i)\$x + input_list[$i+1,$combine_conditional.column_x] ## shifts x coordinates according to tabular file
        coord(msidata_$i)\$y = coord(msidata_$i)\$y + input_list[$i+1,$combine_conditional.column_y] ## shifts y coordinates according to tabular file
        pixel_vector = append(pixel_vector, rep(paste($i+1, input_list[$i+1,$combine_conditional.column_names], sep="_"),times=ncol(msidata_$i))) ## stores file name for each pixel
        msidata_$i\$combined_sample = rep(paste($i+1, input_list[$i+1,$combine_conditional.column_names], sep="_"),times=ncol(msidata_$i))
        pixelcoords_$i = cbind(coord(msidata_$i)[,1:2], rep($i+1,ncol(msidata_$i)))
        #silent $pixelcoords.append('pixelcoords_'+str($i))
        colnames(pixelcoords_$i)[3] = "file_number"

            msidata_$i\$combined_sample = as.factor(names_vector)
            }
        }

        ## Number of input files define grid which is row-wise filled with files

        coord(msidata_$i)\$x = coord(msidata_$i)\$x - (min(coord(msidata_$i)\$x-1)) + x_shifts
        coord(msidata_$i)\$y = coord(msidata_$i)\$y - (min(coord(msidata_$i)\$y-1)) + y_shifts
        x_shifts = max(coord(msidata_$i)\$x) + 5
        max_y = append(max_y, max(coord(msidata_$i)\$y))
        all_files = $num_infiles
        new_row = ($i+1)/ceiling(sqrt(all_files))
        new_row%%1==0

    ## create dataframe with x,y,sample_name and show all pixels in PDF as QC
    position_df = cbind(coord(msidata_combined)[,1:2], msidata_combined\$combined_sample)
    colnames(position_df)[3] = "sample_name"

     ## save as (.RData)

    msidata = msidata_combined
    save(msidata, file="$msidata_combined")

################################## xy shifts ###################################
################################################################################

#elif str( $combine_conditional.combine_method ) == 'xy_shifts':

    ## save as (.RData)

    msidata = msidata_combined
    save(msidata, file="$msidata_combined")

    ## create x,y,sample_name dataframe for QC pdf

    position_df = cbind(coord(msidata), msidata\$combined_sample)
    colnames(position_df)[3] = "sample_name"

        #end if 

    }else{
        print("No annotation tabular output because file has no features or pixels left")
    }


    ]]></configfile>
    </configfiles>
    <inputs>

                <param name="column_y" data_ref="coordinates_file" label="Column with values for shift in y direction" type="data_column"/>
                <param name="column_names" data_ref="coordinates_file" label="Column with dataset names" type="data_column"/>
                <param name="xy_header" type="boolean" label="Tabular files contain a header line" truevalue="TRUE" falsevalue="FALSE"/>
            </when>
        </conditional>
    </inputs>
    <outputs>
        <data format="rdata" name="msidata_combined" label="${tool.name} on ${on_string}"/>
        <data format="pdf" name="QC_overview" from_work_dir="Combined_qc.pdf" label = "${tool.name} on ${on_string}: QC"/>
        <data format="tabular" name="pixel_annotations" label="${tool.name} on ${on_string}: annotations"/>
    </outputs>
    <tests>
        <test>
            <param name="infiles" value="msidata_1.RData,msidata_2.RData,msidata_3.RData" ftype="rdata"/>
            <conditional name="annotation_cond">

**Output**

- single imzML file containing all valid input files
- pdf that shows the pixel positions and annotations of the combined files
- Tabular file with pixel annotations (x,y,column with input file names and up to five annotation columns)


]]>

1:65245dc812c3

<tool id="cardinal_combine" name="MSI combine" version="@VERSION@.1">
    <description>
        combine several mass spectrometry imaging datasets into one
    </description>
    <macros>
        <import>macros.xml</import>
    </macros>
    <expand macro="requirements">
        <requirement type="package" version="3.0">r-ggplot2</requirement>
    </expand>
    <command detect_errors="exit_code">
    <![CDATA[
        #for $i, $infile in enumerate($infiles):
            #if $infile.ext == 'imzml'

                ln -s '$annotation_file' annotation_file_${i}.tabular &&
            #end for
        #end if

        cat '${msi_combine}' &&
        Rscript '${msi_combine}' &&
        #if $imzml_output:
            mkdir $outfile_imzml.files_path &&
            ls -l &&
            mv ./out.imzML "${os.path.join($outfile_imzml.files_path, 'imzml')}" | true &&
            mv ./out.ibd "${os.path.join($outfile_imzml.files_path, 'ibd')}" | true &&
        #end if
        echo "imzML file:" > $outfile_imzml &&
        ls -l "$outfile_imzml.files_path" >> $outfile_imzml

    ]]>
    </command>
    <configfiles>
        <configfile name="msi_combine"><![CDATA[

    ## read and manipulate MSI data

    #if $infile.ext == 'imzml'
        #if str($processed_cond.processed_file) == "processed":
            msidata_$i <- readImzML('infile_${i}', mass.accuracy=$processed_cond.accuracy, units.accuracy = "$processed_cond.units", attach.only=TRUE)
            centroided(msidata_$i) = $centroids
        #else
            msidata_$i <- readImzML('infile_${i}', attach.only=TRUE)
            centroided(msidata_$i) = $centroids
        #end if
    #elif $infile.ext == 'analyze75'
            msidata_$i <- readAnalyze('infile_${i}', attach.only=TRUE)
            centroided(msidata_$i) = $centroids
    #else
            msidata_$i = loadRData('infile_${i}.RData')
    #end if

    ## remove duplicated coordinates, otherwise combine will fail

        msidata_$i\$column3 = rep(NA, ncol(msidata_$i))
        msidata_$i\$column4 = rep(NA, ncol(msidata_$i))
        msidata_$i\$column5 = rep(NA, ncol(msidata_$i))
        msidata_$i\$combined_sample = rep(NA, ncol(msidata_$i))


    #if str($annotation_cond.annotation_tabular) == 'annotation'
        print("annotations")

        ## read annotation tabular, set first two columns as x and y, merge with coordinates dataframe and order according to pixelorder in msidata
        input_annotation = read.delim("annotation_file_${i}.tabular", header = $annotation_cond.tabular_header, stringsAsFactors = FALSE)

        msidata_$i\$column2 = annotation_df_8_sorted[,5]
        msidata_$i\$column3 = annotation_df_8_sorted[,6]
        msidata_$i\$column4 = annotation_df_8_sorted[,7]
        msidata_$i\$column5 = annotation_df_8_sorted[,8]


        ## extract columnnames from (last) annotation tabular (for QC plot names)
        annotation_colnames = colnames(input_annotation)[-c(1,2)]

    #end if

    ################### preparation xy shifts ##########################

    #if str( $combine_conditional.combine_method ) == 'xy_shifts':

        ## shift coordinates according to input tabular file and store file names
        coord(msidata_$i)\$x = as.integer(coord(msidata_$i)\$x + input_list[$i+1,$combine_conditional.column_x]) ## shifts x coordinates according to tabular file
        coord(msidata_$i)\$y = as.integer(coord(msidata_$i)\$y + input_list[$i+1,$combine_conditional.column_y]) ## shifts y coordinates according to tabular file
        pixel_vector = append(pixel_vector, rep(paste($i+1, input_list[$i+1,$combine_conditional.column_names], sep="_"),times=ncol(msidata_$i))) ## stores file name for each pixel
        msidata_$i\$combined_sample = rep(paste($i+1, input_list[$i+1,$combine_conditional.column_names], sep="_"),times=ncol(msidata_$i))
        pixelcoords_$i = cbind(coord(msidata_$i)[,1:2], rep($i+1,ncol(msidata_$i)))
        #silent $pixelcoords.append('pixelcoords_'+str($i))
        colnames(pixelcoords_$i)[3] = "file_number"

            msidata_$i\$combined_sample = as.factor(names_vector)
            }
        }

        ## Number of input files define grid which is row-wise filled with files
        coord(msidata_$i)\$x = as.integer(coord(msidata_$i)\$x - (min(coord(msidata_$i)\$x-1)) + x_shifts)
        coord(msidata_$i)\$y = as.integer(coord(msidata_$i)\$y - (min(coord(msidata_$i)\$y-1)) + y_shifts)
        x_shifts = max(coord(msidata_$i)\$x) + 5
        max_y = append(max_y, max(coord(msidata_$i)\$y))
        all_files = $num_infiles
        new_row = ($i+1)/ceiling(sqrt(all_files))
        new_row%%1==0

    ## create dataframe with x,y,sample_name and show all pixels in PDF as QC
    position_df = cbind(coord(msidata_combined)[,1:2], msidata_combined\$combined_sample)
    colnames(position_df)[3] = "sample_name"

     ## save as (.RData)
    msidata = msidata_combined
    save(msidata, file="$msidata_combined")

    ## save msidata as imzML file
    #if $imzml_output:
        writeImzML(msidata, "out")
    #end if


################################## xy shifts ###################################
################################################################################

#elif str( $combine_conditional.combine_method ) == 'xy_shifts':

    ## save as (.RData)

    msidata = msidata_combined
    save(msidata, file="$msidata_combined")

    ## save msidata as imzML file
    writeImzML(msidata, "out")

    ## create x,y,sample_name dataframe for QC pdf

    position_df = cbind(coord(msidata), msidata\$combined_sample)
    colnames(position_df)[3] = "sample_name"

        #end if 

    }else{
        print("No annotation tabular output because file has no features or pixels left")
    }




    ]]></configfile>
    </configfiles>
    <inputs>

                <param name="column_y" data_ref="coordinates_file" label="Column with values for shift in y direction" type="data_column"/>
                <param name="column_names" data_ref="coordinates_file" label="Column with dataset names" type="data_column"/>
                <param name="xy_header" type="boolean" label="Tabular files contain a header line" truevalue="TRUE" falsevalue="FALSE"/>
            </when>
        </conditional>
        <param name="imzml_output" type="boolean" label="Output of imzML file" truevalue="TRUE" falsevalue="FALSE"/>
    </inputs>
    <outputs>
        <data format="rdata" name="msidata_combined" label="${tool.name} on ${on_string}"/>
        <data format="pdf" name="QC_overview" from_work_dir="Combined_qc.pdf" label = "${tool.name} on ${on_string}: QC"/>
        <data format="tabular" name="pixel_annotations" label="${tool.name} on ${on_string}: annotations"/>
        <data format="imzml" name="outfile_imzml" label="${tool.name} on ${on_string} : imzML">
            <filter>imzml_output</filter>
       </data>
    </outputs>
    <tests>
        <test>
            <param name="infiles" value="msidata_1.RData,msidata_2.RData,msidata_3.RData" ftype="rdata"/>
            <conditional name="annotation_cond">

**Output**

- 
- MSI data as .RData output (can be read with the Cardinal package in R)
- optional: MSI data as imzML file
- pdf that shows the pixel positions and annotations of the combined files
- Tabular file with pixel annotations (x,y,column with input file names and up to five annotation columns)


]]>