comparison preprocessing.xml @ 9:ca727a6dede6 draft

"planemo upload for repository https://github.com/galaxyproteomics/tools-galaxyp/tree/master/tools/cardinal commit 6e8b69ee3aff3c93f745a5de11cc9169130f2e5e"

8:87bb011a4ee8

<tool id="cardinal_preprocessing" name="MSI preprocessing" version="@VERSION@.2">
    <description>
        mass spectrometry imaging preprocessing
    </description>
    <macros>
        <import>macros.xml</import>
    </macros>
    <expand macro="requirements">
        <requirement type="package" version="2.3">r-gridextra</requirement>
        <requirement type="package" version="3.2.1">r-ggplot2</requirement>
    </expand>
    <command detect_errors="exit_code">
    <![CDATA[

        @INPUT_LINKING@

            vectorofactions = append(vectorofactions, "peak binned")
            print(plot(msidata, pixel=random_spectra))
            title("Spectra after peak binning", outer=TRUE, line=0)


    ############################### Data reduction ###########################

        #elif str( $method.methods_conditional.preprocessing_method) == 'Data_reduction':
            print('Data_reduction')

            ## these functions only work on MSImageSet
            msidata = as(msidata, "MSImageSet")

            #if str( $method.methods_conditional.methods_for_reduction.reduction_method) == 'bin':
                print('bin reduction')

                msidata = reduceDimension(msidata, method="bin", width=$method.methods_conditional.methods_for_reduction.bin_width, units="$method.methods_conditional.methods_for_reduction.bin_units", fun=$method.methods_conditional.methods_for_reduction.bin_fun)

                ## optional: replace NA with 0
                #if $method.methods_conditional.methods_for_reduction.replace_NA_bin:
                    ## count and replace NAs
                    print(paste0("Number of NA that were set to zero after binning:",sum(is.na(spectra(msidata)))))
                    spectra(msidata)[is.na(spectra(msidata))] = 0 
                #end if

            #elif str( $method.methods_conditional.methods_for_reduction.reduction_method) == 'resample':
                print('resample reduction')

                msidata = reduceDimension(msidata, method="resample", step=$method.methods_conditional.methods_for_reduction.resample_step)
            #end if

            ## coercition into new format
            msidata = as(msidata, "MSImagingExperiment")

            ############################### QC ###########################

            maxfeatures =nrow(msidata)
            pixelcount = ncol(msidata)
            minmz = round(min(mz(msidata)), digits=2)
            maxmz = round(max(mz(msidata)), digits=2)
            reduced = c(minmz, maxmz,maxfeatures, pixelcount)
            QC_numbers= cbind(QC_numbers, reduced)
            vectorofactions = append(vectorofactions, "reduced")
            print(plot(msidata, pixel=random_spectra))
            title("Spectra after data reduction", outer=TRUE, line=0)

        ############################### Transformation ###########################

        #elif str( $method.methods_conditional.preprocessing_method) == 'Transformation':
            print('Transformation')

                    <option value="mz_alignment">m/z alignment</option>
                    <option value="Peak_picking">Peak picking</option>
                    <option value="Peak_alignment">Peak alignment</option>
                    <option value="Peak_filtering">Peak filtering</option>
                    <option value="Peak_binning">Peak binning to reference peaks</option>
                    <option value="Data_reduction">Data reduction</option>
                    <option value="Transformation">Transformation</option>
                </param>
                <when value="Normalization">
                    <conditional name="methods_for_normalization">
                        <param name="normalization_method" type="select" label="Normalization method">

                           label="Should the peak height or area under the curve be taken as the intensity value?">
                            <option value="height" selected="True">height</option>
                            <option value="area">area</option>
                    </param>
                </when>
                <when value="Data_reduction">
                    <conditional name="methods_for_reduction">
                        <param name="reduction_method" type="select" label="Reduction method">
                            <option value="bin" selected="True">bin</option>
                            <option value="resample">resample</option>
                        </param>
                        <when value="bin">
                            <param name="bin_width" type="float" value="1"
                                   label="The width of a bin in m/z or ppm" help="Width must be greater than range of m/z values divided by number of m/z features"/>
                            <param name="bin_units" type="select" display="radio"
                                   label="Unit for bin">
                                    <option value="mz" selected="True">mz</option>
                                    <option value="ppm">ppm</option>
                            </param>
                            <param name="bin_fun" type="select" display="radio"
                                   label="Calculate sum or mean intensity for ions of the same bin">
                                    <option value="mean" selected="True">mean</option>
                                    <option value="sum">sum</option>
                            </param>
                            <param name="replace_NA_bin" type="boolean" label="Replace NA with 0" truevalue="TRUE" falsevalue="FALSE" checked="True" help="Binning can introduce NAs, should they be replaced with 0"/>
                        </when>
                        <when value="resample">
                            <param name="resample_step" type="float" value="1"
                                   label="The step size in m/z" help="Step size must be greater than range of m/z values divided by number of m/z features"/>
                        </when>
                    </conditional>
                </when>
                <when value="Transformation">
                    <conditional name="transf_conditional">
                        <param name="trans_type" type="select" label="Intensity transformations" help="logarithm base 2 (log2) or squareroot (sqrt)">

                    <param name="preprocessing_method" value="mz_alignment"/>
                </conditional>
            </repeat>
            <repeat name="methods">
                <conditional name="methods_conditional">
                    <param name="preprocessing_method" value="Data_reduction"/>
                    <conditional name="methods_for_reduction">
                        <param name="reduction_method" value="bin"/>
                        <param name="bin_width" value="0.1"/>
                    </conditional>
                </conditional>
            </repeat>
            <output name="QC_overview" file="preprocessing_results4.pdf" compare="sim_size"/>
            <output name="outfile_imzml" ftype="imzml" file="preprocessing_results4.imzml.txt" compare="sim_size">
                <extra_files type="file" file="preprocessing_results4.imzml" name="imzml" lines_diff="6"/>

        </test>
        <test>
            <expand macro="processed_infile_imzml"/>
            <conditional name="processed_cond">
                <param name="processed_file" value="processed"/>
                <param name="accuracy" value="100"/>
                <param name="units" value="ppm"/>
            </conditional>
            <repeat name="methods">
                <conditional name="methods_conditional">
                    <param name="preprocessing_method" value="Transformation"/>

9:ca727a6dede6

<tool id="cardinal_preprocessing" name="MSI preprocessing" version="@VERSION@.0">
    <description>
        mass spectrometry imaging preprocessing
    </description>
    <macros>
        <import>macros.xml</import>
    </macros>
    <expand macro="requirements">
        <requirement type="package" version="2.3">r-gridextra</requirement>
        <requirement type="package" version="3.3.2">r-ggplot2</requirement>
    </expand>
    <command detect_errors="exit_code">
    <![CDATA[

        @INPUT_LINKING@

            vectorofactions = append(vectorofactions, "peak binned")
            print(plot(msidata, pixel=random_spectra))
            title("Spectra after peak binning", outer=TRUE, line=0)


    ############################### Mass binning ###########################

        #elif str( $method.methods_conditional.preprocessing_method) == 'Mass_binning':
            print('mass binning')

            #if str( $method.methods_conditional.mz_range.features_filtering) == 'change_mz_range':

                #if str($processed_cond.processed_file) == "processed":

                msidata = mse_bin = mzBin(msidata,resolution=$method.methods_conditional.bin_width, from=$method.methods_conditional.mz_range.min_mz, to=$method.methods_conditional.mz_range.max_mz, units="$method.methods_conditional.bin_units", fun=$method.methods_conditional.bin_fun)
                
                #else 
                    msidata = mse_bin = mzBin(msidata,resolution=$method.methods_conditional.bin_width, units="$method.methods_conditional.bin_units", fun=$method.methods_conditional.bin_fun)
                #end if

		   
	    #elif str( $method.methods_conditional.mz_range.features_filtering) == 'none':	    

                msidata = mse_bin = mzBin(msidata,resolution=$method.methods_conditional.bin_width, units="$method.methods_conditional.bin_units", fun=$method.methods_conditional.bin_fun)

            #end if
             
            msidata <- process(msidata, BPPARAM=MulticoreParam(workers=number_cpu))
                
                ## optional: replace NA with 0
                #if $method.methods_conditional.replace_NA_bin:
                    ## count and replace NAs
                    print(paste0("Number of NA that were set to zero after binning:",sum(is.na(spectra(msidata)))))
                    spectra(msidata)[is.na(spectra(msidata))] = 0 
                #end if
                
            ############################### QC ###########################

            maxfeatures =nrow(msidata)
            pixelcount = ncol(msidata)
            minmz = round(min(mz(msidata)), digits=2)
            maxmz = round(max(mz(msidata)), digits=2)
            reduced = c(minmz, maxmz,maxfeatures, pixelcount)
            QC_numbers= cbind(QC_numbers, reduced)
            vectorofactions = append(vectorofactions, "reduced")
            print(plot(msidata, pixel=random_spectra))
            title("Spectra after m/z binning", outer=TRUE, line=0)

        ############################### Transformation ###########################

        #elif str( $method.methods_conditional.preprocessing_method) == 'Transformation':
            print('Transformation')

                    <option value="mz_alignment">m/z alignment</option>
                    <option value="Peak_picking">Peak picking</option>
                    <option value="Peak_alignment">Peak alignment</option>
                    <option value="Peak_filtering">Peak filtering</option>
                    <option value="Peak_binning">Peak binning to reference peaks</option>
                    <option value="Mass_binning">m/z binning</option>
                    <option value="Transformation">Transformation</option>
                </param>
                <when value="Normalization">
                    <conditional name="methods_for_normalization">
                        <param name="normalization_method" type="select" label="Normalization method">

                           label="Should the peak height or area under the curve be taken as the intensity value?">
                            <option value="height" selected="True">height</option>
                            <option value="area">area</option>
                    </param>
                </when>
                <when value="Mass_binning">
                    <param name="bin_width" type="float" value="1" label="The width of a bin in m/z or ppm" help="Width must be greater than range of m/z values divided by number of m/z features"/>
		    <param name="bin_units" type="select" display="radio"
		           label="Unit for bin">
		            <option value="mz" selected="True">mz</option>
		            <option value="ppm">ppm</option>
		    </param>
                    <param name="bin_fun" type="select" display="radio"
                           label="Calculate sum or mean intensity for ions of the same bin">
                            <option value="mean" selected="True">mean</option>
                            <option value="sum">sum</option>
                    </param>
                    <param name="replace_NA_bin" type="boolean" label="Replace NA with 0" truevalue="TRUE" falsevalue="FALSE" checked="True" help="Binning can introduce NAs, should they be replaced with 0"/>
                    <conditional name="mz_range">
            		 <param name="features_filtering" type="select" label="Select m/z feature filtering option">
                            <option value="none" selected="True">none</option>
                            <option value="change_mz_range">change m/z range</option>
                        </param>
			    <when value="none"/>
			    <when value="change_mz_range">
				<param name="min_mz" type="float" value="1" label="Minimum value for m/z"/>
				<param name="max_mz" type="float" value="10000" label="Maximum value for m/z"/>
			    </when>
                    </conditional>
                </when>
                <when value="Transformation">
                    <conditional name="transf_conditional">
                        <param name="trans_type" type="select" label="Intensity transformations" help="logarithm base 2 (log2) or squareroot (sqrt)">

                    <param name="preprocessing_method" value="mz_alignment"/>
                </conditional>
            </repeat>
            <repeat name="methods">
                <conditional name="methods_conditional">
                    <param name="preprocessing_method" value="Mass_binning"/>
                        <param name="bin_width" value="0.1"/>
                        <param name="bin_units" value="mz"/>
                </conditional>
            </repeat>
            <output name="QC_overview" file="preprocessing_results4.pdf" compare="sim_size"/>
            <output name="outfile_imzml" ftype="imzml" file="preprocessing_results4.imzml.txt" compare="sim_size">
                <extra_files type="file" file="preprocessing_results4.imzml" name="imzml" lines_diff="6"/>

        </test>
        <test>
            <expand macro="processed_infile_imzml"/>
            <conditional name="processed_cond">
                <param name="processed_file" value="processed"/>
                <param name="accuracy" value="200"/>
                <param name="units" value="ppm"/>
            </conditional>
            <repeat name="methods">
                <conditional name="methods_conditional">
                    <param name="preprocessing_method" value="Transformation"/>