view combine.xml @ 4:48c07268f341 draft

planemo upload for repository https://github.com/galaxyproteomics/tools-galaxyp/tree/master/tools/cardinal commit ecdc3a64aa245d80dbc5487b2bf10a85a43adc6d
author galaxyp
date Fri, 22 Mar 2019 08:18:29 -0400
parents 129cddd02600
children b41107d8fe89
line wrap: on
line source

<tool id="cardinal_combine" name="MSI combine" version="@VERSION@.3">
    <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 $infile.ext == 'imzml'
                ln -s '${infile.extra_files_path}/imzml' infile_${i}.imzML &&
                ln -s '${infile.extra_files_path}/ibd' infile_${i}.ibd &&
            #elif $infile.ext == 'analyze75'
                ln -s '${infile.extra_files_path}/hdr' infile_${i}.hdr &&
                ln -s '${infile.extra_files_path}/img' infile_${i}.img &&
                ln -s '${infile.extra_files_path}/t2m' infile_${i}.t2m &&
            #else
                ln -s '$infile' infile_${i}.RData &&
            #end if
        #end for
        #if $annotation_cond.annotation_tabular == 'annotation'
            #for $i, $annotation_file in enumerate($annotation_cond.annotation_files):
                ln -s '$annotation_file' annotation_file_${i}.tabular &&
            #end for
        #end if

        cat '${msi_combine}' &&
        Rscript '${msi_combine}' &&
        #if str($imzml_output) == "imzml_format":
            mkdir $outfile_imzml.files_path &&
            mv ./out.imzML "${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[

#import re

################ load libraries and some preparations #################

library(Cardinal)
library(ggplot2)

## read tabular file for xy_shift option
#if str( $combine_conditional.combine_method ) == 'xy_shifts':
    input_list = read.delim("$combine_conditional.coordinates_file", header = $combine_conditional.xy_header, 
    stringsAsFactors = FALSE)
#end if

## function to load RData and store with new variable name
loadRData <- function(fileName){
load(fileName)
get(ls()[ls() != "fileName"])
}

## preparations for reading files one by one with for loop
pixel_vector = numeric()
x_shifts = 0
y_shifts = 0
max_y = numeric()
valid_dataset = logical()
#set $msidata = []
#set $pixelcoords = []
#set $num_infiles = len($infiles)
all_files = $num_infiles


############## reading files and changing pixel coordinates ###################

#for $i, $infile in enumerate($infiles):

    ## 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
    print(paste0(sum(duplicated(coord(msidata_$i))), " duplicated coordinates were removed from input file"))
    msidata_${i} <- msidata_${i}[,!duplicated(coord(msidata_${i}))]

    ## same name for MSI data files necessary to combine data in one single coordinate system
    sampleNames(msidata_$i) = "msidata"


    ## read and process annotation tabular or automatically use name of infile as annotation

        ## set all pixel annotations to NA, necessary in case files were combined before with different annotations
        msidata_$i\$column1 = rep(NA, ncol(msidata_$i))
        msidata_$i\$column2 = rep(NA, ncol(msidata_$i))
        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)

        colnames(input_annotation)[1:2] = c("x", "y")
        msidata_coordinates = cbind(coord(msidata_$i)[,1:2], 1:ncol(msidata_$i))
        colnames(msidata_coordinates)[3] = "pixel_index"

        ## only first 5 annotation columns are kept
        if (ncol(input_annotation) > 7){
            input_annotation = input_annotation[,1:7]}

        annotation_df = merge(msidata_coordinates, input_annotation, by=c("x", "y"), all.x=TRUE)
        annotation_df_8 = cbind(annotation_df, data.frame(matrix(NA,ncol=8-ncol(annotation_df), nrow=ncol(msidata_$i)))) 
        annotation_df_8_sorted = annotation_df_8[order(annotation_df_8\$pixel_index),]## orders pixel according to msidata

        ## each annotation column is assigned to the pixel in the pData slot of the MSIdata
        msidata_$i\$column1 = annotation_df_8_sorted[,4]
        msidata_$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_$i))
        ## store number of file
        pixelcoords_$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"

    ################### preparation automatic combination ##########################

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

        ## use name of Galaxy inputfile as combined sample annotation
        names_vector = character()
        #set escaped_element_identifier = re.sub('[^\w\-\s\[/]]', '_', str($infile.element_identifier))

        if (ncol(msidata_$i)>0 & nrow(msidata_$i) >0)
        {
            if (is.null(levels(msidata_$i\$combined_sample)))
            {
            names_vector = append(names_vector, rep(paste($i+1, "$escaped_element_identifier", sep="_"),ncol(msidata_$i)))
            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
        if (new_row%%1==0)
        {x_shifts = 0 ### when row is filled: x values start again at zero
         y_shifts = max(max_y) + 5 ### when row is filled: y value increases to start a new row
        max_y = numeric()}

    #end if

    ## store files to combine them later and for each file check if it is valid

    #silent $msidata.append('msidata_'+str($i))
    valid_dataset = append(valid_dataset, ncol(msidata_$i)>0 & nrow(msidata_$i)>0) ## file with no intensities is considered valid

#end for


###################### automatic combination ###################################
################################################################################

#if str( $combine_conditional.combine_method ) == 'automatic_combine':
    print("automatic_combine")

    ## combine only valid datasets

    valid_data =  list(#echo ','.join($msidata)#)[valid_dataset]
    msidata_combined = do.call(combine, valid_data)
    print("Valid datasets in order of input bottom to top:")
    print(valid_dataset)

    ## 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"

    msidata = msidata_combined

    ## save msidata as imzML file
    #if str($imzml_output) == "imzml_format":
        writeImzML(msidata, "out")
    #elif str($imzml_output) == "rdata_format":
        ## save as (.RData)
        iData(msidata) = iData(msidata)[]
        save(msidata, file="$outfile_rdata")
    #end if


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

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

    ## in case user made mistake with xy shifts: find duplicated coordinates
    all_coordinates = do.call(rbind, list(#echo ','.join($pixelcoords)#))
    duplicated_coordinates= duplicated(all_coordinates[,1:2])| duplicated(all_coordinates[,1:2], fromLast=TRUE)
print(paste0("Number of removed duplicated coordinates after combination: ", sum(duplicated_coordinates)/2))
    unique_coordinates = all_coordinates[!duplicated_coordinates,]

    ## remove duplicated coordinates
    datasetlist = list()
    count = 1
    for (usable_dataset in list(#echo ','.join($msidata)#)){
        pixelsofinterest = pixels(usable_dataset)[names(pixels(usable_dataset)) %in% rownames(unique_coordinates)]
        filtered_dataset = usable_dataset[,pixelsofinterest]
        if (ncol(filtered_dataset) > 0 ){
            datasetlist[[count]] = filtered_dataset}
        count = count +1}

    msidata_combined = do.call(combine, datasetlist)

    msidata = msidata_combined

    ## save msidata as imzML file
    #if str($imzml_output) == "imzml_format":
        writeImzML(msidata, "out")
    #elif str($imzml_output) == "rdata_format":
        ## save as (.RData)
        iData(msidata) = iData(msidata)[]
        save(msidata, file="$outfile_rdata")
    #end if

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

    position_df = cbind(coord(msidata)[,1:2], msidata\$combined_sample)
    colnames(position_df)[3] = "sample_name"

#end if


################################## outputs ####################################
################################################################################

########### QC with pixels and their annotations ################################

pdf("Combined_qc.pdf", width=15, height=15)

## combined plot
combine_plot = ggplot(position_df, aes(x=x, y=y, fill=sample_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)~sample_name, data=position_df, mean)
coord_labels\$file_number = gsub( "_.*$", "", coord_labels\$sample_name)
for(file_count in 1:nrow(coord_labels))
    {combine_plot = combine_plot + annotate("text",x=coord_labels[file_count,"x"],
    y=coord_labels[file_count,"y"],label=toString(coord_labels[file_count,4]))}
print(combine_plot)

    #if str($annotation_cond.annotation_tabular) == 'annotation'
        ## annotation plots

        ## plot 1
        column1_df = cbind(coord(msidata)[,1:2], msidata\$column1)
        colnames(column1_df)[3] = "column1"

        if (sum(is.na(column1_df[3])) < nrow(column1_df)){
        column1_plot = ggplot(column1_df, aes(x=x, y=y, fill=column1))+
               geom_tile() +
               coord_fixed()+
               ggtitle(paste0(annotation_colnames[1]))+
               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, title=annotation_colnames[1]))
        print(column1_plot)}
        ##rename columnname for output tabular file
        colnames(column1_df)[3] = annotation_colnames[1]

        ## plot 2
        column2_df = cbind(coord(msidata)[,1:2], msidata\$column2)
        colnames(column2_df)[3] = "column2"

        if (sum(is.na(column2_df[3])) < nrow(column2_df)){
        column2_plot = ggplot(column2_df, aes(x=x, y=y, fill=column2))+
               geom_tile() +
               coord_fixed()+
               ggtitle(paste0(annotation_colnames[2]))+
               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, title=annotation_colnames[2]))
        print(column2_plot)}

        ##rename columnname for output tabular file
        colnames(column2_df)[3] = annotation_colnames[2]

        ## plot 3
        column3_df = cbind(coord(msidata)[,1:2], msidata\$column3)
        colnames(column3_df)[3] = "column3"
        if (sum(is.na(column3_df[3])) < nrow(column3_df)){
        column3_plot = ggplot(column3_df, aes(x=x, y=y, fill=column3))+
               geom_tile() +
               coord_fixed()+
               ggtitle(paste0(annotation_colnames[3]))+
               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, title=annotation_colnames[3]))
        print(column3_plot)}
        ##rename columnname for output tabular file
        colnames(column3_df)[3] = annotation_colnames[3]

        ## plot 4
        column4_df = cbind(coord(msidata)[,1:2], msidata\$column4)
        colnames(column4_df)[3] = "column4"

        if (sum(is.na(column4_df[3])) < nrow(column4_df)){
        column4_plot = ggplot(column4_df, aes(x=x, y=y, fill=column4))+
               geom_tile() +
               coord_fixed()+
               ggtitle(paste0(annotation_colnames[4]))+
               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, title=annotation_colnames[4]))
        print(column4_plot)}
        ##rename columnname for output tabular file
        colnames(column4_df)[3] = annotation_colnames[4]

        ## plot5

        column5_df = cbind(coord(msidata)[,1:2], msidata\$column5)
        colnames(column5_df)[3] = "column5"
        if (sum(is.na(column5_df[3])) < nrow(column5_df)){
        column5_plot = ggplot(column5_df, aes(x=x, y=y, fill=column5))+
               geom_tile() +
               coord_fixed()+
               ggtitle(paste0(annotation_colnames[5]))+
               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, title=annotation_colnames[5]))
        print(column5_plot)}
        ##rename columnname for output tabular file
        colnames(column5_df)[3] = annotation_colnames[5]

    #end if
    dev.off()

##################### annotation tabular output ################################

    if (length(features(msidata))> 0 & length(pixels(msidata)) > 0){
        position_df\$sample_name = gsub("^[^_]*_","",position_df\$sample_name)

        #if str($annotation_cond.annotation_tabular) == 'no_annotation':

            write.table(position_df, file="$pixel_annotations", quote = FALSE, row.names = FALSE, col.names=TRUE, sep = "\t")

        #else
            annotation_df_list = list(position_df, column1_df, column2_df, column3_df, column4_df, column5_df)
            combined_annotations = Reduce(function(...) merge(..., by=c("x", "y"), all=TRUE), annotation_df_list)
            write.table(combined_annotations, file="$pixel_annotations", quote = FALSE, row.names = FALSE, col.names=TRUE, sep = "\t")
        #end if 

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




    ]]></configfile>
    </configfiles>
    <inputs>
        <param name="infiles" type="data" multiple="true" format="imzml,rdata,analyze75"
            label="MSI data"
            help="Input file as imzML (composite upload), Analyze7.5 (composite upload) or Cardinal MSImageSet saved as RData (regular upload)"/>
        <param name="centroids" type="boolean" label="Centroided input" help="Choose Yes if peak detection has already been done." truevalue="TRUE" falsevalue="FALSE"/>
        <conditional name="processed_cond">
            <param name="processed_file" type="select" label="Processed imzML file" help="Choose no if your input is an Analyze7.5 or continuous imzML file">
                <option value="no_processed" selected="True">no</option>
                <option value="processed">yes</option>
            </param>
            <when value="no_processed"/>
            <when value="processed">
                <param name="accuracy" type="float" value="50" label="Mass accuracy to which the m/z values will be binned" help="This should be set to the native accuracy of the mass spectrometer, if known"/>
                <param name="units" display="radio" type="select" label="Unit of the mass accuracy" help="either m/z or ppm">
                    <option value="mz" >mz</option>
                    <option value="ppm" selected="True" >ppm</option>
                </param>
            </when>
        </conditional>
        <conditional name="annotation_cond">
            <param name="annotation_tabular" type="select" label="Optional annotation of pixels with tabular files">
                <option value="no_annotation" selected="True">no annotation</option>
                <option value="annotation">pixel annotations</option>
            </param>
            <when value="no_annotation"/>
            <when value="annotation">
                <param name="annotation_files" type="data" multiple="true" format="tabular"
                    label="Pixel annotations tabular files"
                    help="Same number and order of files as input files. First column x values, second column y values. Up to 5 columns with pixel annotations"/>
                <param name="tabular_header" type="boolean" label="Tabular files contain a header line" truevalue="TRUE" falsevalue="FALSE"/>
            </when>
        </conditional>
        <conditional name="combine_conditional">
            <param name="combine_method" type="select" label="Way of combining multiple files">
                <option value="automatic_combine" selected="True" >automatic combination</option>
                <option value="xy_shifts">shift xy coordinates with a tabular file</option>
            </param>
            <when value="automatic_combine"/>
            <when value="xy_shifts">
                <param name="coordinates_file" type="data" format="tabular" label="Datasetnames, x and y values to shift pixel coordinates before combining"
            help="Tabular file with three columns: 1 for the filename, 1 for the x-coordinate shift and 1 for the y-coordinate shift. Pixels with the same coordinates after shifting will be deleted."/>
                <param name="column_x" data_ref="coordinates_file" label="Column with values for shift in x direction" type="data_column"/>
                <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="select" display = "radio" optional = "False"
               label="Output format" help= "Choose the output format">
                <option value="imzml_format">imzML</option>
                <option value="rdata_format" selected="True" >RData</option>
        </param>
    </inputs>
    <outputs>
        <data format="imzml" name="outfile_imzml" label="${tool.name} on ${on_string}: imzML">
            <filter>imzml_output=='imzml_format'</filter>
        </data>
        <data format="rdata" name="outfile_rdata" label="${tool.name} on ${on_string}: RData">
            <filter>imzml_output == 'rdata_format'</filter>
       </data>
        <data format="pdf" name="QC_overview" from_work_dir="Combined_qc.pdf" label = "${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">
                <param name="annotation_tabular" value="annotation"/>
                <param name="annotation_files" value="annotations_file1.tabular,annotations_file2.tabular,annotations_file3.tabular" ftype="tabular"/>
                <param name="tabular_header" value="TRUE"/>
            </conditional>
            <param name="combine_method" value="xy_shifts"/>
            <param name="coordinates_file" ftype="tabular" value="xy_coordinates.tabular"/>
            <param name="column_x" value="1"/>
            <param name="column_y" value="2"/>
            <param name="column_names" value="3"/>
            <param name="imzml_output" value="rdata_format"/>
            <output name="pixel_annotations" file="123_annotation_output.tabular"/>
            <output name="outfile_rdata" file="123_combined.RData" compare="sim_size" />
            <output name="QC_overview" file="123_combined_QC.pdf" compare="sim_size"/>
        </test>
        <test>
            <param name="infiles" value="msidata_1.RData,msidata_2.RData,msidata_3.RData" ftype="rdata"/>
            <conditional name="annotation_cond">
                <param name="annotation_tabular" value="no_annotation"/>
            </conditional>
            <param name="combine_method" value="automatic_combine"/>
            <param name="imzml_output" value="imzml_format"/>
            <output name="QC_overview" file="123_combined_auto.pdf" compare="sim_size"/>
            <output name="pixel_annotations" file="123_combined_auto.tabular"/>
            <output name="outfile_imzml" ftype="imzml" file="123_combined_auto.imzml.txt" compare="sim_size">
                <extra_files type="file" file="123_combined_auto.imzml" name="imzml" lines_diff="4"/>
                <extra_files type="file" file="123_combined_auto.ibd" name="ibd" compare="sim_size"/>
            </output>
        </test>
        <test>
            <param name="infiles" value="msidata_1.RData,msidata_2.RData" ftype="rdata"/>
            <conditional name="annotation_cond">
                <param name="annotation_tabular" value="annotation"/>
                <param name="annotation_files" value="annotations_file1.tabular,annotations_file2.tabular" ftype="tabular"/>
                <param name="tabular_header" value="TRUE"/>
            </conditional>
            <param name="combine_method" value="automatic_combine"/>
            <param name="imzml_output" value="rdata_format"/>
            <output name="pixel_annotations" file="12_annotation_output.tabular"/>
            <output name="outfile_rdata" file="12_combined.RData" compare="sim_size" />
            <output name="QC_overview" file="12_combined_QC.pdf" compare="sim_size"/>
        </test>
        <test>
            <param name="infiles" value="msidata_1.RData,123_combined.RData" ftype="rdata"/>
            <conditional name="annotation_cond">
                <param name="annotation_tabular" value="annotation"/>
                <param name="annotation_files" value="annotations_file1.tabular,123_annotation.tabular" ftype="tabular"/>
                <param name="tabular_header" value="TRUE"/>
            </conditional>
            <param name="combine_method" value="automatic_combine"/>
            <param name="imzml_output" value="imzml_format"/>
            <output name="pixel_annotations" file="112_annotation_output.tabular"/>
            <output name="QC_overview" file="112_auto_combined_QC.pdf" compare="sim_size"/>
            <output name="outfile_imzml" ftype="imzml" file="112_auto_combined.imzml.txt" compare="sim_size">
                <extra_files type="file" file="112_auto_combined.imzml" name="imzml" lines_diff="4"/>
                <extra_files type="file" file="112_auto_combined.ibd" name="ibd" compare="sim_size"/>
            </output>
        </test>
        <test>
            <param name="infiles" value="msidata_2.RData,123_combined.RData" ftype="rdata"/>
            <conditional name="annotation_cond">
                <param name="annotation_tabular" value="no_annotation"/>
            </conditional>
            <param name="combine_method" value="automatic_combine"/>
            <param name="imzml_output" value="rdata_format"/>
            <output name="pixel_annotations" file="2123_annotation_output.tabular"/>
            <output name="outfile_rdata" file="2123_auto_combined.RData" compare="sim_size" />
            <output name="QC_overview" file="2123_auto_combined_QC.pdf" compare="sim_size"/>
        </test>
    </tests>
    <help>
<![CDATA[

@CARDINAL_DESCRIPTION@

-----

This tool uses the Cardinal combine function to combine several mass spectrometry imaging data. 

@MSIDATA_INPUT_DESCRIPTION@
- MSI data files must have the same m/z values (to obtain same m/z values for different files: filtering tool same m/z range and preprocessing tool same binning width)
- Coordinates stored as decimals rather than integers will be rounded to obtain a regular pixel grid. This might lead to duplicated coordinates which will be automatically removed before the tools analysis starts. 
@SPECTRA_TABULAR_INPUT_DESCRIPTION@

- For xy shifts with tabular file: Tabular file with x and y coordinates shift and file name

    - Each input file is shifted in x and y direction according to this tabular file. In the example the files have about the same pixel dimensions which is smaller than 510x260.
    - The file can have any column names as header (in this case set "Tabular file contains a header line" to yes) or no header at all

        ::
        
            x_shift     y_shift    file name
               0            0       file1
              510           0       file2
               0           260      file3
              510          260      file4
              ...
              ...
        



**Options**

- "automatic combination": files are automatically arranged in a grid (duplicated pixels are allowed), subfiles are named according to the input file name
- "xy shifts": each file can be moved in x and y direction according to the users need (define one tabular file in the order in which the files are loaded in the history (bottom to top) and define for each file the x and y coordinates shifts in separate columns and the file name in a third column as shown above). The xy shift option combines all datasets and removes all duplicated pixels (same x and y coordinates).

**Tips**

- The combine tools puts all samples into a common x-y-grid, therefore pixel coordinates will change. In case the pixels are already annotated, the annotations should be provided as tabular files and the tool will return an annotation file with the new pixel coordinates. This annotation file can then be used together with the combined MSI data for tools in which the annotation is required (e.g. 'MSI classification') or useful (e.g. 'MSI spectra plots').
- In case more annotations are required: The annotation input file should have an identifier column, for example the patient_ID. A second tabular file that contains more annotations and also one column with the identifier column (e.g. 'patient_ID') can be merged to the annotation output file of this tool with the tool 'join two files' and then set the 'Column to use' parameters for both files to the identifier column. 



**Output**

- MSI data as imzML file or .RData (can be read with the Cardinal package in R)
- 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)


]]>
    </help>
    <expand macro="citations"/>
</tool>