comparison msi_filtering.xml @ 9:28ac8199d4d5 draft default tip

planemo upload for repository https://github.com/galaxyproteomics/tools-galaxyp/tree/master/tools/msi_filtering commit e87eea03505ab1ba067e192bbbcdc6197dc4b42e

8:262db9893c6f

<tool id="mass_spectrometry_imaging_filtering" name="MSI filtering" version="1.10.0.6">
    <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>

    ############ 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)
        colnames(position_df)[3] = "annotation"
        position_df\$annotation = factor(position_df\$annotation)

        ## 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}

    ########### Pixels wihin x and y minima and maxima are kept ###################

    #elif str($pixels_cond.pixel_filtering) == "pixel_range":
        print("pixel range")

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

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



if (npeaks_before_filtering > 0)

{

    #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 #######################

        ## Spectra multiplied with m/z (potential number of peaks)
        numpeaks2 = ncol(spectra(msidata)[])*nrow(spectra(msidata)[])
        ## Percentage of intensities > 0
        percpeaks2 = round(npeaks2/numpeaks2*100, digits=2)
        ## Number of empty TICs
        TICs2 = colSums(spectra(msidata)[], na.rm=TRUE) 
        NumemptyTIC2 = sum(TICs2 == 0)
        ## median TIC
        medint2 = round(median(TICs2), digits=2)

        properties = c("Number of m/z features",

                       "Number of pixels", 
                       "Range of x coordinates", 
                       "Range of y coordinates",
                       "Intensities > 0",
                       "Median TIC per pixel",
                       "Number of zero TICs", 
                       "pixel overview", 
                       "feature overview")

        before = c(paste0(maxfeatures), 
                   paste0(minmz, " - ", maxmz), 

    #if $output_matrix:

        spectramatrix = spectra(msidata)[]
        spectramatrix = cbind(mz(msidata),spectramatrix)
        newmatrix = rbind(c("mz | spectra", names(pixels(msidata))), spectramatrix)
        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")

         <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>
        <test expect_num_outputs="2">

            <param name="annotation_column" value="2"/>
            <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">
                <composite_data value="Example_Continuous.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">
                <composite_data value="Example_Continuous.imzML"/>

            <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>
        <test expect_num_outputs="2">
            <param name="infile" value="" ftype="imzml">
                <composite_data value="Example_Continuous.imzML"/>
                <composite_data value="Example_Continuous.ibd"/>

            <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"/>
            <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">
                <composite_data value="Example_Continuous.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">
                <composite_data value="Analyze75.hdr"/>

            <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>
        <test expect_num_outputs="2">
           <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[

- optional: intensity matrix as tabular file (intensities for m/z in rows and pixel in columns)


Tip: 

- It is recommended to use the filtering tool only for m/z which have been extracted from the same dataset. If you have m/z from dataset A and you want to use them to filter dataset B, first find the corresponding (closest) features in dataset B by using the tool "Join two files on column allowing a small difference". Afterwards use the corresponding feature m/z from dataset A to filter dataset B. 


        ]]>
    </help>
    <citations>

9:28ac8199d4d5

<tool id="mass_spectrometry_imaging_filtering" name="MSI filtering" version="1.10.0.7">
    <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>

    ############ 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
        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)]

        ## rewrite into x = 1, y = 1 format and filter msidata, count validpixels
        pixelvector = character()
        for (pixel in 1:nrow(inputpixels)){
            pixelvector[pixel] = paste0("x = ", inputpixels[pixel,1],", ", "y = ", inputpixels[pixel,2])}
        pixelsofinterest= pixels(msidata)[names(pixels(msidata)) %in% pixelvector]
        msidata = msidata[,pixelsofinterest]
        validpixels=ncol(msidata)

        ## in case some pixels are left print annotation plot
        colnames(inputpixels) = c("x", "y", "annotation")
        position_df = merge(coord(msidata)[,1:2], inputpixels, by=c("x", "y"), all.x=TRUE)
        colnames(position_df)[3] = "annotation"
        position_df\$annotation = factor(position_df\$annotation)


    ########### Pixels wihin x and y minima and maxima are kept ###################

    #elif str($pixels_cond.pixel_filtering) == "pixel_range":
        print("pixel range")

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

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


if (npeaks_before_filtering > 0)

{

    #end if

    ## save msidata as Rfile
    save(msidata, file="$msidata_filtered")
        ## Number of empty TICs
        TICs2 = colSums(spectra(msidata)[], na.rm=TRUE)
}else{
    print("Inputfile or file filtered for pixels has no intensities > 0")
    numberfeatures = NA
    validmz = NA
    ## Number of empty TICs
    TICs2 = NA
}

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

        ## Spectra multiplied with m/z (potential number of peaks)
        numpeaks2 = ncol(spectra(msidata)[])*nrow(spectra(msidata)[])
        ## Percentage of intensities > 0
        percpeaks2 = round(npeaks2/numpeaks2*100, digits=2)
        ## Number of empty TICs
        NumemptyTIC2 = sum(TICs2 == 0)
        ## median TIC
        medint2 = round(median(TICs2), digits=2)

        properties = c("Number of m/z features",

                       "Number of pixels", 
                       "Range of x coordinates", 
                       "Range of y coordinates",
                       "Intensities > 0",
                       "Median TIC per pixel",
                       "Number of empty spectra", 
                       "pixel overview", 
                       "feature overview")

        before = c(paste0(maxfeatures), 
                   paste0(minmz, " - ", maxmz), 

    #if $output_matrix:

        spectramatrix = spectra(msidata)[]
        spectramatrix = cbind(mz(msidata),spectramatrix)
        newmatrix = rbind(c("mz", names(pixels(msidata))), spectramatrix)
        write.table(newmatrix, file="$intensity_matrix", quote = FALSE, row.names = FALSE, col.names=FALSE, sep = "\t")

    #end if

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

         <param name="output_matrix" type="boolean" display="radio" label="Intensity matrix output"/>
    </inputs>

    <outputs>
        <data format="rdata" name="msidata_filtered" label="$infile.display_name filtered"/>
        <data format="pdf" name="QC_overview" from_work_dir="filtertool_QC.pdf" label = "$infile.display_name filtered_QC"/>
        <data format="tabular" name="intensity_matrix" label="$infile.display_name filtered_matrix">
            <filter>output_matrix</filter>
        </data>
    </outputs>
    <tests>
        <test expect_num_outputs="2">

            <param name="annotation_column" value="2"/>
            <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="QC_overview" 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">
                <composite_data value="Example_Continuous.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="QC_overview" 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">
                <composite_data value="Example_Continuous.imzML"/>

            <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="QC_overview" file="imzml_filtered3.pdf" compare="sim_size" delta="20000"/>
            <output name="msidata_filtered" file="imzml_filtered3.RData" compare="sim_size"/>
            <output name="intensity_matrix" file="imzml_matrix3.tabular"/>
        </test>
        <test expect_num_outputs="2">
            <param name="infile" value="" ftype="imzml">
                <composite_data value="Example_Continuous.imzML"/>
                <composite_data value="Example_Continuous.ibd"/>

            <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"/>
            <output name="QC_overview" 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">
                <composite_data value="Example_Continuous.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="QC_overview" 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">
                <composite_data value="Analyze75.hdr"/>

            <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="QC_overview" file="analyze_filtered.pdf" compare="sim_size" delta="20000"/>
            <output name="msidata_filtered" file="analyze_filtered.RData" compare="sim_size" />
            <output name="intensity_matrix" file="analyze_matrix.tabular"/>
        </test>
        <test expect_num_outputs="2">
           <param name="infile" value="" ftype="analyze75">
                <composite_data value="Analyze75.hdr"/>
                <composite_data value="Analyze75.img"/>
                <composite_data value="Analyze75.t2m"/>
            </param>
            <output name="QC_overview" 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="intensity_matrix" file="rdata_matrix.tabular"/>
            <output name="msidata_filtered" file="rdata_notfiltered.RData" compare="sim_size" />
            <output name="QC_overview" file="rdata_notfiltered.pdf" compare="sim_size" />
        </test>
    </tests>
    <help>
        <![CDATA[

- optional: intensity matrix as tabular file (intensities for m/z in rows and pixel in columns)


Tip: 

- It is recommended to use the filtering tool only for m/z which have been extracted from the same dataset. If the m/z values are from a different dataset, the tool "Join two files on column allowing a small difference" should be used to find corresponding m/z values, which can then be used for filtering. 


        ]]>
    </help>
    <citations>