annotate msi_combine.xml @ 1:f3f6c32ab690 draft

planemo upload for repository https://github.com/galaxyproteomics/tools-galaxyp/tree/master/tools/msi_combine commit dd64f41874a56c4e2619bf58ae3681d806cf9b3f

<tool id="mass_spectrometry_imaging_combine" name="MSI combine" version="1.10.0.0">
    <description>
        combine several mass spectrometry imaging datasets into one
    </description>
    <requirements>
        <requirement type="package" version="1.10.0">bioconductor-cardinal</requirement>
        <requirement type="package" version="2.2.1">r-ggplot2</requirement>
    </requirements>
    <command detect_errors="exit_code">
    <![CDATA[
        #for $i, $infile in enumerate($infiles):
            #if $infile.ext == 'imzml'
                ln -s '${infile.extra_files_path}/imzml' infile.imzML &&
                ln -s '${infile.extra_files_path}/ibd' infile.ibd &&
            #elif $infile.ext == 'analyze75'
                ln -s '${infile.extra_files_path}/hdr' infile.hdr &&
                ln -s '${infile.extra_files_path}/img' infile.img &&
                ln -s '${infile.extra_files_path}/t2m' infile.t2m &&
            #else
                ln -s '$infile' infile_${i}.RData &&
            #end if
        #end for
        cat '${msi_combine}' &&
        Rscript '${msi_combine}'

    ]]>
    </command>
    <configfiles>
        <configfile name="msi_combine"><![CDATA[
#import re
################ load libraries, read rename and combine files #################

library(Cardinal)
library(ggplot2)

#if str( $combine_conditional.combine_method ) == 'xy_shifts':
    input_list = read.delim("$combine_conditional.coordinates_file", header = FALSE, 
    stringsAsFactors = FALSE)
#end if

pixel_vector = numeric()
names_vector = character()
x_shifts = 0
y_shifts = 0
max_y = numeric()

#set $msidata = []
#set $pixelcoords = []
#set $num_infiles = len($infiles)

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

    #if $infile.ext == 'imzml'
        msidata_$i <- readImzML('infile_${i}')
        sampleNames(msidata_$i) = "msidata"
        pixelcoords_$i = cbind(coord(msidata_$i)[,1:2], rep($i+1,ncol(msidata_$i)))
    #elif $infile.ext == 'analyze75'
        msidata_$i <- readAnalyze('infile_${i}')
        sampleNames(msidata_$i) = "msidata"
        pixelcoords_$i = cbind(coord(msidata_$i)[,1:2], rep($i+1,ncol(msidata_$i)))
    #else
        loadRData <- function(fileName){
        #loads an RData file, and returns it
        load(fileName)
        get(ls()[ls() != "fileName"])
        }
        msidata_$i = loadRData('infile_${i}.RData')
        sampleNames(msidata_$i) = "msidata"
        pixelcoords_$i = cbind(coord(msidata_$i)[,1:2], rep($i+1,ncol(msidata_$i)))
    #end if
        colnames(pixelcoords_$i)[3] = "file_number"

    #if str( $combine_conditional.combine_method ) == 'xy_shifts':
        coord(msidata_$i)\$x = coord(msidata_$i)\$x + input_list[$i+1,$combine_conditional.column_x]
        coord(msidata_$i)\$y = coord(msidata_$i)\$y + input_list[$i+1,$combine_conditional.column_y]
        pixel_vector = append(pixel_vector, rep(input_list[$i+1,$combine_conditional.column_names],times=ncol(msidata_$i)))
    
    #elif str( $combine_conditional.combine_method ) == 'automatic_combine':
        #set escaped_element_identifier = re.sub('[^\w\-\s\[/]]', '_', str($infile.element_identifier))
        names_vector = append(names_vector, rep(paste($i+1, "$escaped_element_identifier", sep="_"),ncol(msidata_$i)))
        coord(msidata_$i)\$x = coord(msidata_$i)\$x - (min(coord(msidata_$i)\$x-1)) + x_shifts
        coord(msidata_$i)\$y = coord(msidata_$i)\$y - (min(coord(msidata_$i)\$y-1)) + y_shifts
        x_shifts = max(coord(msidata_$i)\$x) + $combine_conditional.x_distance
        max_y = append(max_y, max(coord(msidata_$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 ### x values start again at zero
         y_shifts = max(max_y) + $combine_conditional.y_distance
        max_y = numeric()}

    #end if
    #silent $msidata.append('msidata_'+str($i))
    #silent $pixelcoords.append('pixelcoords_'+str($i))
#end for

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

#if str( $combine_conditional.combine_method ) == 'automatic_combine':
    print("automatic_combine")
    msidata_combined = do.call(combine, list(#echo ','.join($msidata)#))
    sample_names = as.factor(names_vector)
    pData(msidata_combined)\$sample = sample_names

   ## create PDF to show pixels of each file
    pdf("combining_qc.pdf", width=15, height=15)
    position_df = cbind(coord(msidata_combined)[,1:2], pData(msidata_combined)\$sample)
    colnames(position_df)[3] = "sample_name"

    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)
    dev.off()

     ## save as (.RData)
    msidata = msidata_combined
    save(msidata, file="$msidata_combined")
################################## xy shifts ###################################
################################################################################
#elif str( $combine_conditional.combine_method ) == 'xy_shifts':
    print("xy_shifts")
    msidata_combined = do.call(combine, list(#echo ','.join($msidata)#))

############# replace NA with 0 and rename pixels according to dataset #########

    spectra(msidata_combined)[is.na(spectra(msidata_combined))] <- 0
    print(paste0("Number of NAs which were replaced ",sum(is.na(spectra(msidata_combined)))))

    sample_names = as.factor(pixel_vector)
    pData(msidata_combined)\$sample = sample_names

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

    ## create PDF to show pixels of each file
    pdf("combining_qc.pdf")
    position_df = cbind(coord(msidata), pData(msidata)\$sample)
    colnames(position_df)[3] = "sample_name"

    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=12))

    coord_labels = aggregate(cbind(x,y)~sample_name, data=position_df, mean)
    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,"sample_name"]))
    }
        
    print(combine_plot)

    dev.off()


################################## no shifts ###################################
################################################################################
#elif str( $combine_conditional.combine_method ) == 'no_shifts':
    print("no_shifts")
    msidata_combined = do.call(combine, list(#echo ','.join($msidata)#))

############# replace NA with 0 and rename pixels according to dataset #########

    spectra(msidata_combined)[is.na(spectra(msidata_combined))] <- 0
    print(paste0("Number of NAs which were replaced ",sum(is.na(spectra(msidata_combined)))))

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

    ## create PDF to show pixels of each file
    pdf("combining_qc.pdf")
    position_df = cbind(coord(msidata), pData(msidata)\$sample)
    colnames(position_df)[3] = "sample_name"

           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=12))

    dev.off()


################################## no combination ##############################
################################################################################
#elif str( $combine_conditional.combine_method ) == 'no_combine':
    print("no_combine")

    ## create PDF to show pixels of each file
    pdf("combining_qc.pdf")
    position_df = do.call(rbind, list(#echo ','.join($pixelcoords)#))
    position_df[duplicated(position_df[,1:2]),3] = 0
    position_df\$file_number = as.factor(position_df\$file_number)

    combine_plot = ggplot(position_df, aes(x=x, y=y, fill=file_number))+
           geom_tile() +
           coord_fixed()+
           ggtitle("Spatial orientation before combination")+
           theme_bw()+
           theme(text=element_text(family="ArialMT", face="bold", size=12))+
           theme(panel.grid.major = element_line(colour = "black")) +
           scale_x_continuous(minor_breaks = seq(min(position_df\$x-50), max(position_df\$x+50, 1))) +
           scale_y_continuous(minor_breaks = seq(min(position_df\$y-50), max(position_df\$y+50, 1)))


    coord_labels = aggregate(cbind(x,y)~file_number, data=position_df, mean)
    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,"file_number"]))
    }
        
    print(combine_plot)

    dev.off()
#end if

#if $output_matrix:

    if (length(features(msidata_combined))> 0 & length(pixels(msidata_combined)) > 0)
    {
        spectramatrix = spectra(msidata_combined)
        rownames(spectramatrix) = mz(msidata_combined)
        newmatrix = rbind(pixels(msidata_combined), spectramatrix)
        write.table(newmatrix[2:nrow(newmatrix),], file="$matrixasoutput", quote = FALSE, row.names = TRUE, col.names=NA, sep = "\t")
    }else{
        print("file has no features or pixels left")
    }
#end if

    ]]></configfile>
    </configfiles>
    <inputs>
        <param name="infiles" type="data" multiple="true" format="imzml,rdata,analyze75"
            label="MSI rawdata as imzml, analyze7.5 or Cardinal MSImageSet saved as RData"
            help="load imzml and ibd file by uploading composite datatype imzml"/>
        <conditional name="combine_conditional">
            <param name="combine_method" type="select" label="Select the way you want to combine multiple files" help="More detailed help can be found in the help section at the bottom">
                <option value="automatic_combine" selected="True" >automatic combination</option>
                <option value="no_shifts" >no coordinates shift</option>
                <option value="xy_shifts">xy shifts by hand</option>
                <option value="no_combine">check pixels before combination</option>
            </param>
            <when value="no_shifts">
            </when>
            <when value="automatic_combine">
                <param name="x_distance" type="integer" value="10" label="How many pixels in x direction should be between files?"/>
                <param name="y_distance" type="integer" value="10" label="How many pixels in y direction should be between files?"/>
            </when>
            <when value="xy_shifts">
                <param name="coordinates_file" type="data" format="tabular" label="datasetnames, X and y values to shift data 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"/>
                <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"/>
            </when>
            <when value="no_combine"/>
        </conditional>
    <param name="output_matrix" type="boolean" display="radio" label="Intensity matrix output"/>
    </inputs>
    <outputs>
        <data format="rdata" name="msidata_combined" label="Combined MSI data"/>
        <data format="pdf" name="combining_qc" from_work_dir="combining_qc.pdf" label = "Combined image of pixels"/>
        <data format="tabular" name="matrixasoutput" label="Combined matrix">
            <filter>output_matrix</filter>
        </data>
    </outputs>
    <tests>
        <test expect_num_outputs="3">
            <param name="infiles" value="msidata_1.RData,msidata_2.RData,msidata_3.RData" ftype="rdata"/>
            <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="output_matrix" value="True"/>
            <output name="matrixasoutput" file="123_combined_matrix.tabular"/>
            <output name="msidata_combined" file="123_combined.RData" compare="sim_size" />
            <output name="combining_qc" file="123_combined_QC.pdf" compare="sim_size" delta="20000"/>
        </test>
        <test expect_num_outputs="3">
            <param name="infiles" value="msidata_1.RData,msidata_2.RData" ftype="rdata"/>
            <param name="combine_method" value="no_shifts"/>
            <param name="output_matrix" value="True"/>
            <output name="matrixasoutput" file="12_combined_matrix.tabular"/>
            <output name="msidata_combined" file="12_combined.RData" compare="sim_size" />
            <output name="combining_qc" file="12_combined_QC.pdf" compare="sim_size" delta="20000"/>
        </test>
        <test expect_num_outputs="3">
            <param name="infiles" value="msidata_1.RData,msidata_2.RData" ftype="rdata"/>
            <param name="combine_method" value="automatic_combine"/>
            <param name="x_distance" value="1"/>
            <param name="y_distance" value="1"/>
            <param name="output_matrix" value="True"/>
            <output name="matrixasoutput" file="12_auto_combined_matrix.tabular"/>
            <output name="msidata_combined" file="12_auto_combined.RData" compare="sim_size" />
            <output name="combining_qc" file="12_auto_combined_QC.pdf" compare="sim_size" delta="20000"/>
        </test>
    </tests>
    <help>
<![CDATA[
This tool can combine several mass-spectrometry imaging files. 
    1) m/z values need to be the same across all datasets
    2) pixels (defined by x and y coordinates) must be unique

1) Same m/z values/axis can be achieved with the filtering tool to get all datasets to the same m/z range and afterwards binning in the preprocessing tool to obtain the same bins for all dataset. 
2) The pixels (defined by x and y coordinates) must be unique across all datasets, therefore the option "Select the way you want to combine multiple files" is helpful:

    - "automatic combination": files are arranged in a grid with a distance in x and y direction which can be given by the user
    - "no coordinates shift": this option can only be used if all pixels are unique across datasets
    - "xy shifts by hand": 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
    - "check pixels before combination": no combination takes place. You will only get a pdf which shows the arrangement of the pixels (with or without additional xy shifts)

Input data: 3 types of input data can be used:

- imzml file (upload imzml and ibd file via the "composite" function) `Introduction to the imzml format <http://ms-imaging.org/wp/introduction/>`_
- Analyze7.5 (upload hdr, img and t2m file via the "composite" function)
- Cardinal "MSImageSet" data (with variable name "msidata", saved as .RData)

The output of this tool is a Cardinal "MSImageSet" with the variable name "msidata" saved as .RData. 
]]>
    </help>
    <citations>
        <citation type="doi">10.1093/bioinformatics/btv146</citation>
    </citations>
</tool>