comparison msi_filtering.xml @ 6:bab12ded74a5 draft

planemo upload for repository https://github.com/galaxyproteomics/tools-galaxyp/tree/master/tools/msi_filtering commit 5bceedc3a11c950790692a4c64bbb83d46897bee

5:3d5ac78fb2b0

<tool id="mass_spectrometry_imaging_filtering" name="MSI filtering" version="1.10.0.3">
    <description>tool for filtering mass spectrometry imaging data</description>
    <requirements>
        <requirement type="package" version="1.10.0">bioconductor-cardinal</requirement>
        <requirement type="package" version="2.2.1">r-gridextra</requirement>
    </requirements>
    <command detect_errors="exit_code">
    <![CDATA[

        #if $infile.ext == 'imzml'

################################# load libraries and read file #################


library(Cardinal)
library(gridExtra)


#if $infile.ext == 'imzml'
    #if str($processed_cond.processed_file) == "processed":
        msidata <- readImzML('infile', mass.accuracy=$processed_cond.accuracy, units.accuracy = "$processed_cond.units")

#else
    load('infile.RData')
#end if


########################### optional QC numbers ########################

if (sum(spectra(msidata)[]>0, na.rm=TRUE) > 0)
{
    #if $outputs.outputs_select == "quality_control":

        ## Number of features (m/z)
        maxfeatures = length(features(msidata))
        ## Range m/z
        minmz = round(min(mz(msidata)), digits=2)

        ## median TIC
        medint = round(median(TICs), digits=2)
        ## Store features for QC plot
        featuresinfile = mz(msidata)

    #end if

    ###################################### Filtering of pixels #####################
    ################################################################################

    #################### Pixels in the one column format "x=,y=" #####################

    #if str($pixels_cond.pixel_filtering) == "single_column":
        print("single column")

        input_list = read.delim("$pixels_cond.single_pixels", header = FALSE, stringsAsFactors = FALSE)
        numberpixels = length(input_list[,$pixels_cond.pixel_column])
        valid_entries = input_list[,$pixels_cond.pixel_column] %in% names(pixels(msidata))
        validpixels = sum(valid_entries)

        if (validpixels != 0){
            pixelsofinterest = pixels(msidata)[names(pixels(msidata)) %in% input_list[valid_entries,$pixels_cond.pixel_column]]
            msidata = msidata[,pixelsofinterest]
        }else{
            msidata = msidata[,0]
            validpixels=0}

    ############ Pixels in two columns format: x and y in different columns #############

    #elif str($pixels_cond.pixel_filtering) == "two_columns":
        print("two columns")

        input_list = read.delim("$pixels_cond.two_columns_pixel", header = FALSE, 
        stringsAsFactors = FALSE)
        numberpixels = length(input_list[,$pixels_cond.pixel_column_x])

        inputpixel_x = input_list[,$pixels_cond.pixel_column_x]
        inputpixel_y = input_list[,$pixels_cond.pixel_column_y]
        inputpixels = cbind(inputpixel_x, inputpixel_y)
        colnames(inputpixels) = c("x", "y")
        valid_rows = merge(inputpixels, coord(msidata)[,1:2])
        validpixels = nrow(valid_rows)

        if (validpixels != 0){
            pixelvector = character()
            for (pixel in 1:nrow(valid_rows)){
                pixelvector[pixel] = paste0("x = ", valid_rows[pixel,1],", ", "y = ", valid_rows[pixel,2])}
            pixelsofinterest= pixels(msidata)[names(pixels(msidata)) %in% pixelvector]
            msidata = msidata[,pixelsofinterest]
        }else{
            validpixels=0}

        print("pixel range")

        numberpixels = "range"
        validpixels = "range"

    ## only filter pixels if at least one pixel will be left

        if (sum(coord(msidata)\$x <= $pixels_cond.max_x_range & coord(msidata)\$x >= $pixels_cond.min_x_range) > 0 & sum(coord(msidata)\$y <= $pixels_cond.max_y_range & coord(msidata)\$y >= $pixels_cond.min_y_range) > 0){

            msidata = msidata[, coord(msidata)\$x <= $pixels_cond.max_x_range & coord(msidata)\$x >= $pixels_cond.min_x_range]
            msidata = msidata[, coord(msidata)\$y <= $pixels_cond.max_y_range & coord(msidata)\$y >= $pixels_cond.min_y_range]
        }else{

        numberpixels = 0
        validpixels = 0

    #end if


}else{
    print("Inputfile has no intensities > 0")

}

###################################### filtering of features ######################
##################################################################################

######################## Keep m/z from tabular file #########################

if (sum(spectra(msidata)[], na.rm=TRUE) > 0){

    #if str($features_cond.features_filtering) == "features_list":
        print("feature list")

        input_features = read.delim("$inputfeatures", header = FALSE, stringsAsFactors = FALSE)
        startingrow = $features_cond.feature_header+1
        extracted_features = input_features[startingrow:nrow(input_features),$features_cond.feature_column]
        numberfeatures = length(extracted_features)

        if (grepl("m/z = ", input_features[startingrow,$features_cond.feature_column])==FALSE){

    ### if input is in numeric format
            if (class(extracted_features) == "numeric"){
                ### max digits given in the input file will be used to match m/z
                max_digits = max(nchar(matrix(unlist(strsplit(as.character(extracted_features), "\\.")), ncol=2, byrow=TRUE)[,2]))
                validfeatures = extracted_features %in% round(mz(msidata),max_digits)
                featuresofinterest = features(msidata)[round(mz(msidata), digits = max_digits) %in% extracted_features[validfeatures]]
                validmz = length(unique(featuresofinterest))
            }else{
                    validmz = 0
                    featuresofinterest = 0}

    ### filter msidata for valid features

        msidata = msidata[featuresofinterest,]

    ############### features within a given range are kept #########################

    #elif str($features_cond.features_filtering) == "features_range":
        print("feature range")

        numberfeatures = "range"

    #end if

    ## save msidata as Rfile
    save(msidata, file="$msidata_filtered")

    #################### optional QC numbers #######################

    #if $outputs.outputs_select == "quality_control":

        ## Number of features (m/z)
        maxfeatures2 = length(features(msidata))
        ## Range m/z
        minmz2 = round(min(mz(msidata)), digits=2)

                   paste0("valid pixels: ", validpixels),
                   paste0("valid mz: ", validmz))

        property_df = data.frame(properties, before, filtered)

    ############################### optional PDF QC ################################

        pdf("filtertool_QC.pdf", fonts = "Times", pointsize = 12)
        plot(0,type='n',axes=FALSE,ann=FALSE)
        title(main=paste0("Qualitycontrol of filtering tool for file: \n\n", "$infile.display_name"))
        grid.table(property_df, rows= NULL)

        ### heatmap image as visual pixel control
        if (length(features(msidata))> 0 & length(pixels(msidata)) > 0){
            image(msidata, mz=$outputs.inputmz, plusminus = $outputs.plusminus_dalton, contrast.enhance = "none", 
              main= paste0($outputs.inputmz," ± ", $outputs.plusminus_dalton, " Da"), ylim = c(maximumy2+0.2*maximumy2,minimumy2-0.2*minimumy2))

            ### control features which are removed
            hist(mz(msidata), xlab="m/z", main="Kept m/z values")
            #if str($features_cond.features_filtering) == "none":
            print("no difference histogram as no m/z filtering took place")
            #else:

                if (isTRUE(all.equal(featuresinfile, mz(msidata)))){
                print("No difference in m/z values before and after filtering, no histogram drawn")
                }else{
                hist(setdiff(featuresinfile, mz(msidata)), xlab="m/z", main="Removed m/z values")}
            #end if
        }else{
            print("file has no features or pixels left")}

        dev.off()

    #end if

    ############################### optional intensity matrix ######################

    #if $output_matrix:

        write.table(newmatrix, file="$matrixasoutput", quote = FALSE, row.names = FALSE, col.names=FALSE, sep = "\t")

    #end if

}else{
    print("Inputfile or file filtered for pixels has no intensities > 0")
}
    ]]></configfile>
    </configfiles>
    <inputs>
        <param name="infile" type="data" format="imzml,rdata,analyze75"

                    <option value="mz" >mz</option>
                    <option value="ppm" selected="True" >ppm</option>
                </param>
            </when>
        </conditional>
        <conditional name="pixels_cond">
            <param name="pixel_filtering" type="select" label="Select pixel filtering option">
                <option value="none" selected="True">none</option>
                <option value="single_column">tabular file with single column (x = 1, y = 1)</option>
                <option value="two_columns">tabular file with separate columns for x and y values</option>

            <when value="none"/>
            <when value="single_column">
                <param name="single_pixels" type="data" format="tabular" label="Pixels in single column for filtering of MSI data"
                    help="tabular file with pixels of interest in the form x = 1, y = 1"/>
                <param name="pixel_column" data_ref="single_pixels" label="Column with pixels" type="data_column"/>
            </when> 
            <when value="two_columns">
                <param name="two_columns_pixel" type="data" format="tabular" label="Pixels in two columns for filtering of MSI data"
                    help="tabular file with pixels of interest in two separate columns"/>
                <param name="pixel_column_x" data_ref="two_columns_pixel" label="Column with x values" type="data_column"/>
                <param name="pixel_column_y" data_ref="two_columns_pixel" label="Column with y values" type="data_column"/>
            </when> 
            <when value="pixel_range">
                <param name="min_x_range" type="integer" value="0" label="Minimum value for x"/>
                <param name="max_x_range" type="integer" value="100" label="Maximum value for x"/>
                <param name="min_y_range" type="integer" value="0" label="Minimum value for y"/>
                <param name="max_y_range" type="integer" value="100" label="Maximum value for y"/>
            </when> 
        </conditional>
        <conditional name="features_cond">
            <param name="features_filtering" type="select" label="Select feature filtering option">
                <option value="none" selected="True">none</option>
                <option value="features_list">keep features (tabular input)</option>
                <option value="features_range">keep features within a range (manual input)</option>

                <param name="removal_plusminus" type="float" value="20" label="Window in which m/z will be removed" help="This value will be added and substracted from the given input value"/>
                <param name="units_removal" type="select" display = "radio" optional = "False" label="units">
                        <option value="ppm" selected="True">ppm</option>
                        <option value="Da">Da</option>
                </param>
            </when> 
        </conditional>
        <conditional name="outputs">
           <param name="outputs_select" type="select" label="Quality control output">
               <option value="quality_control" selected="True">yes</option>
               <option value="no_quality_control">no</option>
           </param>
           <when value="quality_control">
              <param name="inputmz" type="float" value="1296.7" label="M/z for which a heatmap image will be drawn" help="Use a m/z which is still present in all pixels to control if the pixel filtering went well"/>
              <param name="plusminus_dalton" value="0.25" type="float" label="Range for m/z value" help="plusminus m/z window"/>
           </when>
           <when value="no_quality_control"/>
         </conditional>
         <param name="output_matrix" type="boolean" display="radio" label="Intensity matrix output"/>
    </inputs>
    <outputs>
        <data format="rdata" name="msidata_filtered" label="$infile.display_name filtered"/>
        <data format="pdf" name="filtering_qc" from_work_dir="filtertool_QC.pdf" label = "$infile.display_name filtered_QC">
            <filter>outputs["outputs_select"] == "quality_control"</filter>
        </data>
        <data format="tabular" name="matrixasoutput" label="$infile.display_name filtered_matrix">
            <filter>output_matrix</filter>
        </data>
    </outputs>
    <tests>

            <param name="pixel_column" value="1"/>
            <param name="features_filtering" value="features_list"/>
            <param name="inputfeatures" ftype="tabular" value = "inputfeatures.tabular"/>
            <param name="feature_column" value="2"/>
            <param name="feature_header" value="1"/>
             <param name="outputs_select" value="quality_control"/>
                <param name="inputmz" value="328.9"/>
                <param name="plusminus_dalton" value="0.25"/>
            <output name="filtering_qc" file="imzml_filtered.pdf" compare="sim_size" delta="20000"/>
            <output name="msidata_filtered" file="imzml_filtered.RData" compare="sim_size" />
        </test>
        <test expect_num_outputs="2">
            <param name="infile" value="" ftype="imzml">

            <param name="pixel_filtering" value="pixel_range"/>
            <param name="min_x_range" value="10"/>
            <param name="max_x_range" value="20"/>
            <param name="min_y_range" value="2"/>
            <param name="max_y_range" value="2"/>
             <param name="outputs_select" value="quality_control"/>
                <param name="inputmz" value="328.9"/>
                <param name="plusminus_dalton" value="0.25"/>
            <output name="filtering_qc" file="imzml_filtered2.pdf" compare="sim_size" delta="20000"/>
            <output name="msidata_filtered" file="imzml_filtered2.RData" compare="sim_size" />
        </test>
        <test expect_num_outputs="3">
            <param name="infile" value="" ftype="imzml">

            <param name="min_y_range" value="2"/>
            <param name="max_y_range" value="2"/>
            <param name="features_filtering" value="features_range"/>
            <param name="min_mz" value="350" />
            <param name="max_mz" value="500"/>
             <param name="outputs_select" value="quality_control"/>
                <param name="inputmz" value="328.9"/>
                <param name="plusminus_dalton" value="0.25"/>
            <param name="output_matrix" value="True"/>
            <output name="filtering_qc" file="imzml_filtered3.pdf" compare="sim_size" delta="20000"/>
            <output name="msidata_filtered" file="imzml_filtered3.RData" compare="sim_size" />
            <output name="matrixasoutput" file="imzml_matrix3.tabular"/>
        </test>

            </param>
            <param name="pixel_filtering" value="two_columns"/>
            <param name="two_columns_pixel" ftype="tabular" value = "inputpixels_2column.tabular"/>
            <param name="pixel_column_x" value="1"/>
            <param name="pixel_column_y" value="3"/>
            <param name="features_filtering" value="features_list"/>
            <param name="inputfeatures" ftype="tabular" value = "inputcalibrantfile2.txt"/>
            <param name="feature_column" value="1"/>
            <param name="feature_header" value="0"/>
             <param name="outputs_select" value="quality_control"/>
                <param name="inputmz" value="328.9"/>
                <param name="plusminus_dalton" value="0.25"/>
            <output name="filtering_qc" file="imzml_filtered4.pdf" compare="sim_size" delta="20000"/>
            <output name="msidata_filtered" file="imzml_filtered4.RData" compare="sim_size" />
        </test>
        <test expect_num_outputs="2">
            <param name="infile" value="" ftype="imzml">

            <param name="max_y_range" value="20"/>
            <param name="features_filtering" value="features_list"/>
            <param name="inputfeatures" ftype="tabular" value = "featuresofinterest5.tabular"/>
            <param name="feature_column" value="1"/>
            <param name="feature_header" value="0"/>
             <param name="outputs_select" value="quality_control"/>
                <param name="inputmz" value="328.9"/>
                <param name="plusminus_dalton" value="0.25"/>
            <output name="filtering_qc" file="imzml_filtered5.pdf" compare="sim_size" delta="20000"/>
            <output name="msidata_filtered" file="imzml_filtered5.RData" compare="sim_size" />
        </test>
        <test expect_num_outputs="3">
           <param name="infile" value="" ftype="analyze75">

            <param name="single_pixels" ftype="tabular" value = "inputpixels2.tabular"/>
            <param name="pixel_column" value="1"/>
            <param name="features_filtering" value="features_list"/>
            <param name="inputfeatures" ftype="tabular" value = "featuresofinterest2.tabular"/>
            <param name="feature_column" value="1"/>
            <conditional name="outputs">
                <param name="outputs_select" value="quality_control"/>
                    <param name="inputmz" value="1200"/>
                    <param name="plusminus_dalton" value="0.25"/>
            </conditional>
            <param name="output_matrix" value="True"/>
            <output name="filtering_qc" file="analyze_filtered.pdf" compare="sim_size" delta="20000"/>
            <output name="msidata_filtered" file="analyze_filtered.RData" compare="sim_size" />
            <output name="matrixasoutput" file="analyze_matrix.tabular"/>
        </test>

           <param name="infile" value="" ftype="analyze75">
                <composite_data value="Analyze75.hdr"/>
                <composite_data value="Analyze75.img"/>
                <composite_data value="Analyze75.t2m"/>
            </param>
            <conditional name="outputs">
                <param name="outputs_select" value="quality_control"/>
                    <param name="inputmz" value="1200"/>
                    <param name="plusminus_dalton" value="0.25"/>
            </conditional>
            <output name="filtering_qc" file="analyze75_filtered2.pdf" compare="sim_size" delta="20000"/>
            <output name="msidata_filtered" file="analyze_filteredoutside.RData" compare="sim_size" />
        </test>
        <test expect_num_outputs="2">
            <param name="infile" value="preprocessed.RData" ftype="rdata"/>
            <conditional name="outputs">
                <param name="outputs_select" value="no_quality_control"/>
            </conditional>
            <param name="output_matrix" value="True"/>
            <output name="matrixasoutput" file="rdata_matrix.tabular"/>
            <output name="msidata_filtered" file="rdata_notfiltered.RData" compare="sim_size" />
        </test>
    </tests>
    <help>
        <![CDATA[

- Cardinal "MSImageSet" data (with variable name "msidata", saved as .RData)


Options:

- pixel filtering: can use a tabular file containing x and y coordinates or by defining a range for x and y by hand
- m/z feature filtering: can use a tabular file containing m/z of interest or by defining a range for the m/z values (! numeric input will be rounded to 2 digits before matching to m/z!)
- m/z feature removing: infering m/z such as matrix contaminants can be removed by specifying their m/z in a tabular file and optionally set a window (window in ppm or m/z in which peaks should be removed)


Output: 

- imzML file filtered for pixels and/or m/z

6:bab12ded74a5

<tool id="mass_spectrometry_imaging_filtering" name="MSI filtering" version="1.10.0.4">
    <description>tool for filtering mass spectrometry imaging data</description>
    <requirements>
        <requirement type="package" version="1.10.0">bioconductor-cardinal</requirement>
        <requirement type="package" version="2.2.1">r-gridextra</requirement>
        <requirement type="package" version="2.2.1">r-ggplot2</requirement>
    </requirements>
    <command detect_errors="exit_code">
    <![CDATA[

        #if $infile.ext == 'imzml'

################################# load libraries and read file #################


library(Cardinal)
library(gridExtra)
library(ggplot2)


#if $infile.ext == 'imzml'
    #if str($processed_cond.processed_file) == "processed":
        msidata <- readImzML('infile', mass.accuracy=$processed_cond.accuracy, units.accuracy = "$processed_cond.units")

#else
    load('infile.RData')
#end if


########################### QC numbers ########################

        ## Number of features (m/z)
        maxfeatures = length(features(msidata))
        ## Range m/z
        minmz = round(min(mz(msidata)), digits=2)

        ## median TIC
        medint = round(median(TICs), digits=2)
        ## Store features for QC plot
        featuresinfile = mz(msidata)

## Next steps will only run if there are more than 0 intensities/pixels/features in the file

if (sum(spectra(msidata)[]>0, na.rm=TRUE) > 0)
{


        ## prepare dataframe for QC of pixel distribution (will be overwritten in filtering of pixels condition)
        position_df = cbind(coord(msidata)[,1:2], rep("$infile.element_identifier", times=ncol(msidata)))


    ###################################### Filtering of pixels #####################
    ################################################################################

    #################### Pixels in the one column format "x=,y=" #####################

    #if str($pixels_cond.pixel_filtering) == "single_column":
        print("single column")

        ## read tabular file, count number of rows (= number of pixels), count how many pixels are valid
        input_list = read.delim("$pixels_cond.single_pixels", header = FALSE, stringsAsFactors = FALSE)
        startingrow = $pixels_cond.pixel_header+1
        numberpixels = length(startingrow:nrow(input_list))
        valid_entries = input_list[startingrow:nrow(input_list),$pixels_cond.pixel_column] %in% names(pixels(msidata))
        validpixels = sum(valid_entries)
        valid_annotations = input_list[valid_entries,c($pixels_cond.pixel_column, $pixels_cond.annotation_column)]

        ## for valid pixels: filter file for pixels and create dataframe with x,y,annotation for QC
        if (validpixels != 0){
            ## filter file for pixels
            pixelsofinterest = pixels(msidata)[names(pixels(msidata)) %in% valid_annotations[,1]]
            msidata = msidata[,pixelsofinterest]
            ## position_df for QC
            pixel_coords = coord(msidata)[names(pixels(msidata)) %in% valid_annotations[,1],1:2]
            position_df = cbind(pixel_coords, valid_annotations[,2])
        }else{
            msidata = msidata[,0]
            validpixels=0}

    ############ Pixels in two columns format: x and y in different columns #############

    #elif str($pixels_cond.pixel_filtering) == "two_columns":
        print("two columns")

        ## read tabular file, count number of rows (= number of pixels), extract dataframe with x,y,annotation (for QC), count number of valid pixels
        input_list = read.delim("$pixels_cond.two_columns_pixel", header = FALSE, 
        stringsAsFactors = FALSE)
        startingrow = $pixels_cond.pixel_header+1
        numberpixels = length(startingrow:nrow(input_list))
        inputpixels = input_list[startingrow:nrow(input_list),c($pixels_cond.pixel_column_x, $pixels_cond.pixel_column_y, $pixels_cond.annotation_column_xy)]
        colnames(inputpixels) = c("x", "y", "annotation")
        position_df = merge(coord(msidata)[,1:2], inputpixels, by=c("x", "y"), all.x=TRUE)

        validpixels = nrow(position_df)

        ## for valid pixels: filter file for pixels
        if (validpixels != 0){
            pixelvector = character()
            for (pixel in 1:nrow(position_df)){
                pixelvector[pixel] = paste0("x = ", position_df[pixel,1],", ", "y = ", position_df[pixel,2])}
            pixelsofinterest= pixels(msidata)[names(pixels(msidata)) %in% pixelvector]
            msidata = msidata[,pixelsofinterest]
        }else{
            validpixels=0}

        print("pixel range")

        numberpixels = "range"
        validpixels = "range"

        ## only filter pixels if at least one pixel will be left
        if (sum(coord(msidata)\$x <= $pixels_cond.max_x_range & coord(msidata)\$x >= $pixels_cond.min_x_range) > 0 & sum(coord(msidata)\$y <= $pixels_cond.max_y_range & coord(msidata)\$y >= $pixels_cond.min_y_range) > 0){

            msidata = msidata[, coord(msidata)\$x <= $pixels_cond.max_x_range & coord(msidata)\$x >= $pixels_cond.min_x_range]
            msidata = msidata[, coord(msidata)\$y <= $pixels_cond.max_y_range & coord(msidata)\$y >= $pixels_cond.min_y_range]
        }else{

        numberpixels = 0
        validpixels = 0

    #end if

}else{
    print("Inputfile has no intensities > 0")
}

    ################################# filtering of features ######################
    ##############################################################################

    ####################### Keep m/z from tabular file #########################

## feature filtering only when pixels/features/intensities are left
if (sum(spectra(msidata)[], na.rm=TRUE) > 0)
{

    #if str($features_cond.features_filtering) == "features_list":
        print("feature list")

        ## read tabular file, define starting row, extract and count valid features
        input_features = read.delim("$inputfeatures", header = FALSE, stringsAsFactors = FALSE)
        startingrow = $features_cond.feature_header+1
        extracted_features = input_features[startingrow:nrow(input_features),$features_cond.feature_column]
        numberfeatures = length(extracted_features)

        ## find out type of tabular file (numeric or character format)
        if (grepl("m/z = ", input_features[startingrow,$features_cond.feature_column])==FALSE){

    ### if input is in numeric format
            if (class(extracted_features) == "numeric"){
                ### max digits given in the input file will be used to match m/z but the maximum is 4
                max_digits = max(nchar(matrix(unlist(strsplit(as.character(extracted_features), "\\.")), ncol=2, byrow=TRUE)[,2]))
                if (max_digits >4)
                {
                    max_digits = 4
                }

                validfeatures = round(extracted_features, max_digits) %in% round(mz(msidata),max_digits)
                featuresofinterest = features(msidata)[round(mz(msidata), digits = max_digits) %in% round(extracted_features[validfeatures], max_digits)]
                validmz = length(unique(featuresofinterest))
            }else{
                    validmz = 0
                    featuresofinterest = 0}

    ### filter msidata for valid features

        msidata = msidata[featuresofinterest,]

    ############### features within a given range are kept #####################

    #elif str($features_cond.features_filtering) == "features_range":
        print("feature range")

        numberfeatures = "range"

    #end if

    ## save msidata as Rfile
    save(msidata, file="$msidata_filtered")

}else{
    print("Inputfile or file filtered for pixels has no intensities > 0")
    numberfeatures = NA
    validmz = NA
}

    #################### QC numbers #######################


        ## Number of features (m/z)
        maxfeatures2 = length(features(msidata))
        ## Range m/z
        minmz2 = round(min(mz(msidata)), digits=2)

                   paste0("valid pixels: ", validpixels),
                   paste0("valid mz: ", validmz))

        property_df = data.frame(properties, before, filtered)

    ############################### PDF QC ################################


        pdf("filtertool_QC.pdf", fonts = "Times", pointsize = 12)
        plot(0,type='n',axes=FALSE,ann=FALSE)
        title(main=paste0("Qualitycontrol of filtering tool for file: \n\n", "$infile.display_name"))
        grid.table(property_df, rows= NULL)

## QC report with more than value-table: only when pixels/features/intensities are left
if (sum(spectra(msidata)[], na.rm=TRUE) > 0)
{
        ### visual pixel control

            colnames(position_df)[3] = "annotation_name"
            pixel_image = ggplot(position_df, aes(x=x, y=y, fill=annotation_name))+
                   geom_tile() +
                   coord_fixed()+
                   ggtitle("Spatial orientation of combined data")+
                   theme_bw()+
                   theme(text=element_text(family="ArialMT", face="bold", size=15))+
                   theme(legend.position="bottom",legend.direction="vertical")+
                   guides(fill=guide_legend(ncol=4,byrow=TRUE))
            coord_labels = aggregate(cbind(x,y)~annotation_name, data=position_df, mean, na.rm=TRUE, na.action="na.pass")
            coord_labels\$file_number = gsub( "_.*$", "", coord_labels\$annotation_name)

            print(pixel_image)

            ### control features which are removed
            hist(mz(msidata), xlab="m/z", main="Kept m/z values")
            #if str($features_cond.features_filtering) == "none":
                print("no difference histogram as no m/z filtering took place")
            #else:

                if (isTRUE(all.equal(featuresinfile, mz(msidata)))){
                print("No difference in m/z values before and after filtering, no histogram drawn")
                }else{
                hist(setdiff(featuresinfile, mz(msidata)), xlab="m/z", main="Removed m/z values")}
            #end if

        dev.off()

    ############################### optional intensity matrix ######################

    #if $output_matrix:

        write.table(newmatrix, file="$matrixasoutput", quote = FALSE, row.names = FALSE, col.names=FALSE, sep = "\t")

    #end if

}else{
    print("Inputfile or filtered file has no intensities > 0")
    dev.off()
}
    ]]></configfile>
    </configfiles>
    <inputs>
        <param name="infile" type="data" format="imzml,rdata,analyze75"

                    <option value="mz" >mz</option>
                    <option value="ppm" selected="True" >ppm</option>
                </param>
            </when>
        </conditional>

        <conditional name="pixels_cond">
            <param name="pixel_filtering" type="select" label="Select pixel filtering option">
                <option value="none" selected="True">none</option>
                <option value="single_column">tabular file with single column (x = 1, y = 1)</option>
                <option value="two_columns">tabular file with separate columns for x and y values</option>

            <when value="none"/>
            <when value="single_column">
                <param name="single_pixels" type="data" format="tabular" label="Pixels in single column for filtering of MSI data"
                    help="tabular file with pixels of interest in the form x = 1, y = 1"/>
                <param name="pixel_column" data_ref="single_pixels" label="Column with pixels" type="data_column"/>
                <param name="annotation_column" data_ref="single_pixels" label="Column with annotations for each pixel" type="data_column"/>
                <param name="pixel_header" label="Number of header lines to skip" value="0" type="integer"/>
            </when> 
            <when value="two_columns">
                <param name="two_columns_pixel" type="data" format="tabular" label="Pixels in two columns for filtering of MSI data"
                    help="tabular file with pixels of interest in two separate columns"/>
                <param name="pixel_column_x" data_ref="two_columns_pixel" label="Column with x values" type="data_column"/>
                <param name="pixel_column_y" data_ref="two_columns_pixel" label="Column with y values" type="data_column"/>
                <param name="annotation_column_xy" data_ref="two_columns_pixel" label="Column with annotations" type="data_column"/>
                <param name="pixel_header" label="Number of header lines to skip" value="0" type="integer"/>
            </when> 
            <when value="pixel_range">
                <param name="min_x_range" type="integer" value="0" label="Minimum value for x"/>
                <param name="max_x_range" type="integer" value="100" label="Maximum value for x"/>
                <param name="min_y_range" type="integer" value="0" label="Minimum value for y"/>
                <param name="max_y_range" type="integer" value="100" label="Maximum value for y"/>
            </when> 
        </conditional>

        <conditional name="features_cond">
            <param name="features_filtering" type="select" label="Select feature filtering option">
                <option value="none" selected="True">none</option>
                <option value="features_list">keep features (tabular input)</option>
                <option value="features_range">keep features within a range (manual input)</option>

                <param name="removal_plusminus" type="float" value="20" label="Window in which m/z will be removed" help="This value will be added and substracted from the given input value"/>
                <param name="units_removal" type="select" display = "radio" optional = "False" label="units">
                        <option value="ppm" selected="True">ppm</option>
                        <option value="Da">Da</option>
                </param>
            </when>
        </conditional>
         <param name="output_matrix" type="boolean" display="radio" label="Intensity matrix output"/>
    </inputs>

    <outputs>
        <data format="rdata" name="msidata_filtered" label="$infile.display_name filtered"/>
        <data format="pdf" name="filtering_qc" from_work_dir="filtertool_QC.pdf" label = "$infile.display_name filtered_QC"/>
        <data format="tabular" name="matrixasoutput" label="$infile.display_name filtered_matrix">
            <filter>output_matrix</filter>
        </data>
    </outputs>
    <tests>

            <param name="pixel_column" value="1"/>
            <param name="features_filtering" value="features_list"/>
            <param name="inputfeatures" ftype="tabular" value = "inputfeatures.tabular"/>
            <param name="feature_column" value="2"/>
            <param name="feature_header" value="1"/>
            <output name="filtering_qc" file="imzml_filtered.pdf" compare="sim_size" delta="20000"/>
            <output name="msidata_filtered" file="imzml_filtered.RData" compare="sim_size" />
        </test>
        <test expect_num_outputs="2">
            <param name="infile" value="" ftype="imzml">

            <param name="pixel_filtering" value="pixel_range"/>
            <param name="min_x_range" value="10"/>
            <param name="max_x_range" value="20"/>
            <param name="min_y_range" value="2"/>
            <param name="max_y_range" value="2"/>
            <output name="filtering_qc" file="imzml_filtered2.pdf" compare="sim_size" delta="20000"/>
            <output name="msidata_filtered" file="imzml_filtered2.RData" compare="sim_size" />
        </test>
        <test expect_num_outputs="3">
            <param name="infile" value="" ftype="imzml">

            <param name="min_y_range" value="2"/>
            <param name="max_y_range" value="2"/>
            <param name="features_filtering" value="features_range"/>
            <param name="min_mz" value="350" />
            <param name="max_mz" value="500"/>
            <param name="output_matrix" value="True"/>
            <output name="filtering_qc" file="imzml_filtered3.pdf" compare="sim_size" delta="20000"/>
            <output name="msidata_filtered" file="imzml_filtered3.RData" compare="sim_size" />
            <output name="matrixasoutput" file="imzml_matrix3.tabular"/>
        </test>

            </param>
            <param name="pixel_filtering" value="two_columns"/>
            <param name="two_columns_pixel" ftype="tabular" value = "inputpixels_2column.tabular"/>
            <param name="pixel_column_x" value="1"/>
            <param name="pixel_column_y" value="3"/>
            <param name="annotation_column_xy" value="2"/>
            <param name="features_filtering" value="features_list"/>
            <param name="inputfeatures" ftype="tabular" value = "inputcalibrantfile2.txt"/>
            <param name="feature_column" value="1"/>
            <param name="feature_header" value="0"/>
            <output name="filtering_qc" file="imzml_filtered4.pdf" compare="sim_size" delta="20000"/>
            <output name="msidata_filtered" file="imzml_filtered4.RData" compare="sim_size" />
        </test>
        <test expect_num_outputs="2">
            <param name="infile" value="" ftype="imzml">

            <param name="max_y_range" value="20"/>
            <param name="features_filtering" value="features_list"/>
            <param name="inputfeatures" ftype="tabular" value = "featuresofinterest5.tabular"/>
            <param name="feature_column" value="1"/>
            <param name="feature_header" value="0"/>
            <output name="filtering_qc" file="imzml_filtered5.pdf" compare="sim_size" delta="20000"/>
            <output name="msidata_filtered" file="imzml_filtered5.RData" compare="sim_size" />
        </test>
        <test expect_num_outputs="3">
           <param name="infile" value="" ftype="analyze75">

            <param name="single_pixels" ftype="tabular" value = "inputpixels2.tabular"/>
            <param name="pixel_column" value="1"/>
            <param name="features_filtering" value="features_list"/>
            <param name="inputfeatures" ftype="tabular" value = "featuresofinterest2.tabular"/>
            <param name="feature_column" value="1"/>
            <param name="output_matrix" value="True"/>
            <output name="filtering_qc" file="analyze_filtered.pdf" compare="sim_size" delta="20000"/>
            <output name="msidata_filtered" file="analyze_filtered.RData" compare="sim_size" />
            <output name="matrixasoutput" file="analyze_matrix.tabular"/>
        </test>

           <param name="infile" value="" ftype="analyze75">
                <composite_data value="Analyze75.hdr"/>
                <composite_data value="Analyze75.img"/>
                <composite_data value="Analyze75.t2m"/>
            </param>
            <output name="filtering_qc" file="analyze75_filtered2.pdf" compare="sim_size" delta="20000"/>
            <output name="msidata_filtered" file="analyze_filteredoutside.RData" compare="sim_size" />
        </test>
        <test expect_num_outputs="3">
            <param name="infile" value="preprocessed.RData" ftype="rdata"/>
            <conditional name="outputs">
                <param name="outputs_select" value="no_quality_control"/>
            </conditional>
            <param name="output_matrix" value="True"/>
            <output name="matrixasoutput" file="rdata_matrix.tabular"/>
            <output name="msidata_filtered" file="rdata_notfiltered.RData" compare="sim_size" />
            <output name="filtering_qc" file="rdata_notfiltered.pdf" compare="sim_size" />
        </test>
    </tests>
    <help>
        <![CDATA[

- Cardinal "MSImageSet" data (with variable name "msidata", saved as .RData)


Options:

- pixel filtering/annotation: either with a tabular file containing x and y coordinates and pixel annotations or by defining a range for x and y by hand (for the latter no annotation is possible)
- m/z feature filtering: can use a tabular file containing m/z of interest or by defining a range for the m/z values (! numeric input will be rounded to 2 digits before matching to m/z!)
- m/z feature removing: perturbing m/z such as matrix contaminants can be removed by specifying their m/z in a tabular file and optionally set a window (window in ppm or m/z in which peaks should be removed)


Output: 

- imzML file filtered for pixels and/or m/z