Mercurial > repos > galaxyp > maldi_quant_preprocessing
diff maldi_quant_preprocessing.xml @ 0:e2aa05746a69 draft
planemo upload for repository https://github.com/galaxyproteomics/tools-galaxyp/tree/master/tools/MALDIquant commit 5feaf3d0e0da8cef1241fecc1f4d6f81324594e6
| author | galaxyp |
|---|---|
| date | Wed, 22 Aug 2018 11:49:06 -0400 |
| parents | |
| children | 0892a051eb17 |
line wrap: on
line diff
--- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/maldi_quant_preprocessing.xml Wed Aug 22 11:49:06 2018 -0400 @@ -0,0 +1,509 @@ +<tool id="maldi_quant_preprocessing" name="MALDIquant preprocessing" version="1.18.0.0"> + <description> + Preprocessing of mass-spectrometry imaging data + </description> + <macros> + <import>maldi_macros.xml</import> + </macros> + <expand macro="requirements"/> + <command detect_errors="exit_code"> + <![CDATA[ + #if $infile.ext == 'imzml' + cp '${infile.extra_files_path}/imzml' infile.imzML && + cp '${infile.extra_files_path}/ibd' infile.ibd && + #elif $infile.ext == 'analyze75' + cp '${infile.extra_files_path}/hdr' infile.hdr && + cp '${infile.extra_files_path}/img' infile.img && + cp '${infile.extra_files_path}/t2m' infile.t2m && + du infile.hdr && + du infile.img && + du -s -B1 infile.hdr && + #else + ln -s $infile infile.RData && + #end if + Rscript "${maldi_quant_preprocessing}" && + 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 && + echo "imzML file:" > $outfile_imzml && + ls -l "$outfile_imzml.files_path" >> $outfile_imzml + ]]> + </command> + <configfiles> + <configfile name="maldi_quant_preprocessing"><![CDATA[ + +@R_IMPORTS@ + +#if $restriction_conditional.restriction == 'restrict': + + print('Reading mask region') + ## Import imzML file + + coordinate_matrix = as.matrix(read.delim("$restriction_conditional.coordinates_file", header = FALSE, stringsAsFactors = FALSE))[,1:2] + + maldi_data = importImzMl('infile.imzML', + coordinates = coordinate_matrix) + pixelnames = paste0("x = ", coordinates(maldi_data)[,1],", y = ", coordinates(maldi_data)[,2]) + +#else: + + print('Reading entire file') + #if $infile.ext == 'imzml' + ## Import imzML file + maldi_data = import( 'infile.imzML', type="imzML" ) + #elif $infile.ext == 'analyze75' + ## Import analyze7.5 file + maldi_data = import( 'infile.hdr' ) + #else + loadRData <- function(fileName){ + #loads an RData file, and returns it + load(fileName) + get(ls()[ls() != "fileName"]) + } + msidata = loadRData('infile.RData') + + ## save coordinates + cardinal_coordinates = as.matrix(Cardinal::coord(msidata)[,1:2]) + ## save mz values + cardinal_mzs = Cardinal::mz(msidata) + ## create MALDIquant MassSpectrum object + maldi_data = list() + for(number_spectra in 1:ncol(msidata)){ + maldi_data[[number_spectra]] = createMassSpectrum(mass = cardinal_mzs, intensity = iData(msidata)[,number_spectra]) + } + + #end if + +#end if + +## Quality control plots during preprocessing + +pdf("prepro_qc_plot.pdf", fonts = "Times", pointsize = 12) +plot(0,type='n',axes=FALSE,ann=FALSE) + +## if no filename is given, name of file in Galaxy history is used + #set $filename = $infile.display_name +title(main=paste("$filename")) + +#if str($tabular_annotation.load_annotation) == 'yes_annotation': + print("use annotation file") + ## read and extract x,y,annotation information + input_tabular = read.delim("$tabular_annotation.annotation_file", header = $tabular_annotation.tabular_header, stringsAsFactors = FALSE) + annotation_input = input_tabular[,c($tabular_annotation.column_x, $tabular_annotation.column_y, $tabular_annotation.column_names)] + colnames(annotation_input) = c("x", "y", "annotation") ## rename annotations header to default name "annotation" + + ## merge with coordinate information of MSI data + coordinates_st = cbind(coordinates(maldi_data)[,1:2], c(1:length(maldi_data))) + colnames(coordinates_st)[3] = "pixel_index" + merged_annotation = merge(coordinates_st, annotation_input, by=c("x", "y"), all.x=TRUE) + merged_annotation[is.na(merged_annotation)] = "NA" + merged_annotation = merged_annotation[order(merged_annotation\$pixel_index),] + samples = as.factor(merged_annotation\$annotation) + +## print annotation overview into PDF output + + ## the more annotation groups a file has the smaller will be the legend + number_combined = length(levels(as.factor(merged_annotation\$annotation))) + if (number_combined<20){ + legend_size = 10 + }else if (number_combined>20 && number_combined<40){ + legend_size = 9 + }else if (number_combined>40 && number_combined<60){ + legend_size = 8 + }else if (number_combined>60 && number_combined<100){ + legend_size = 7 + }else{ + legend_size = 6 + } + + combine_plot = ggplot(merged_annotation, aes(x=x, y=y, fill=annotation))+ + geom_tile() + + coord_fixed()+ + ggtitle("Spatial orientation of annotated data")+ + theme_bw()+ + theme(plot.title = element_text(hjust = 0.5))+ + theme(text=element_text(family="ArialMT", face="bold", size=12))+ + theme(legend.position="bottom",legend.direction="vertical")+ + theme(legend.key.size = unit(0.2, "line"), legend.text = element_text(size = legend_size))+ + guides(fill=guide_legend(ncol=5,byrow=TRUE)) + + print(combine_plot) + +#end if + +#################### Preprocessing methods ##################################### + +## QC plot +avgSpectra = averageMassSpectra(maldi_data,method="mean") +plot(avgSpectra, main="Average spectrum for input file") + +#for $method in $methods: + + #if str( $method.methods_conditional.method ) == 'Transformation': + + print('transforming') + ##transformation + maldi_data = transformIntensity(maldi_data, method="$method.methods_conditional.transform_method") + ## QC plot + avgSpectra = averageMassSpectra(maldi_data,method="mean") + plot(avgSpectra, main="Average spectrum after transformation") + + + #elif str( $method.methods_conditional.method ) == 'Smoothing': + + print('smoothing') + ##smoothing + + #if str($method.methods_conditional.methods_for_smoothing.smooth_method ) == 'SavitzkyGolay': + print('SavitzkyGolay') + + maldi_data = smoothIntensity(maldi_data, + method="SavitzkyGolay", polynomialOrder=$method.methods_conditional.methods_for_smoothing.polynomial, + halfWindowSize=$method.methods_conditional.halfWindowSize) + + #elif str($method.methods_conditional.methods_for_smoothing.smooth_method ) == 'MovingAverage': + print('MovingAverage') + + maldi_data = smoothIntensity(maldi_data, + method="MovingAverage", weighted=$method.methods_conditional.methods_for_smoothing.weighted, + halfWindowSize=$method.methods_conditional.halfWindowSize) + + #end if + + ## QC plot + avgSpectra = averageMassSpectra(maldi_data,method="mean") + plot(avgSpectra, main="Average spectrum after smoothing") + + + #elif str( $method.methods_conditional.method ) == 'Baseline': + + print('baseline removing') + ## Remove baseline + + maldi_data = removeBaseline(maldi_data, + method="$method.methods_conditional.baseline_method", + iterations=$method.methods_conditional.iterations) + ## QC plot + avgSpectra = averageMassSpectra(maldi_data,method="mean") + plot(avgSpectra, main="Average spectrum after baseline removal") + + + #elif str( $method.methods_conditional.method ) == 'Calibrate': + + print('calibrate') + ##calibrate + + #if $method.methods_conditional.mass_start != 0 and $method.methods_conditional.mass_end != 0: + ## calibrate only given m/z range + maldi_data = calibrateIntensity(maldi_data, + method="$method.methods_conditional.calibrate_method", + range=c($method.methods_conditional.mass_start, $method.methods_conditional.mass_end)) + #else: + maldi_data = calibrateIntensity(maldi_data, + method="$method.methods_conditional.calibrate_method") + #end if + ## QC plot + avgSpectra = averageMassSpectra(maldi_data,method="mean") + plot(avgSpectra, main="Average spectrum after normalization") + + + #elif str( $method.methods_conditional.method ) == 'Align': + + print('align') + ##align spectra + + #if str($method.methods_conditional.reference_for_alignment.align_ref) == 'no_reference': + + maldi_data = alignSpectra(maldi_data, halfWindowSize=$method.methods_conditional.halfWindowSize, + SNR=$method.methods_conditional.snr, + tolerance=$method.methods_conditional.tolerance, + warpingMethod="$method.methods_conditional.warping_method") + + #elif str($method.methods_conditional.reference_for_alignment.align_ref) == 'yes_reference': + + ## create reference mass_vector from tabular file + mass_vector = read.delim("$method.methods_conditional.reference_for_alignment.reference_file", header = FALSE, stringsAsFactors = FALSE)[,1] + int_vector = rep(1,length(mass_vector)) + mass_list = createMassPeaks(mass_vector, int_vector) + + maldi_data = alignSpectra(maldi_data, halfWindowSize=$method.methods_conditional.halfWindowSize, + SNR=$method.methods_conditional.snr, + tolerance=$method.methods_conditional.tolerance, + warpingMethod="$method.methods_conditional.warping_method", + reference = mass_list, allowNoMatches =$method.methods_conditional.reference_for_alignment.allow_nomatch, emptyNoMatches = $method.methods_conditional.reference_for_alignment.empty_nomatch) + + #if $method.methods_conditional.reference_for_alignment.remove_empty: + + #if $infile.ext == 'rdata' + cardinal_coordinates = cardinal_coordinates[-findEmptyMassObjects(maldi_data),] ## remove coordinates of empty spectra for Cardinal RData input + #end if + #if str($tabular_annotation.load_annotation) == 'yes_annotation': + merged_annotation = merged_annotation[-findEmptyMassObjects(maldi_data),] ## remove coordinate annotations for empty spectra + #end if + maldi_data = removeEmptyMassObjects(maldi_data) + #end if + #end if + + ## QC plot + + if (length(maldi_data)>0){ + avgSpectra = averageMassSpectra(maldi_data,method="mean") + plot(avgSpectra, main="Average spectrum after alignment") + }else{"All spectra are empty"} + + #end if +#end for + +dev.off() + +## export imzML file +if (length(maldi_data)>0){ + #if $infile.ext == 'rdata' + MALDIquantForeign::exportImzMl(maldi_data, file="out.imzMl", processed=$export_processed, coordinates=cardinal_coordinates) + #else + MALDIquantForeign::exportImzMl(maldi_data, file="out.imzMl", processed=$export_processed) + #end if + + ## export annotation tabular file + #if str($tabular_annotation.load_annotation) == 'yes_annotation': + write.table(merged_annotation, file="$annotation_output", quote = FALSE, row.names = FALSE, col.names=TRUE, sep = "\t") + #end if +}else{"All spectra are empty, outputfiles will be empty,too."} + + ]]> + </configfile> + </configfiles> + <inputs> + <param name="infile" type="data" format="imzml,rdata" label="MS metadata" help="This file is in imzML format or Cardinal MSImageSet saved as RData"/> + <conditional name="restriction_conditional"> + <param name="restriction" type="select" label="Read in only spectra of interest" help="This option only works for imzML files"> + <option value="no_restriction" selected="True">Calculate on entire file</option> + <option value="restrict">Restrict to coordinates of interest</option> + </param> + <when value="restrict"> + <param name="coordinates_file" type="data" format="tabular" label="Tabular file with coordinates which should be read" help="x-values in first column, y-values in second column"/> + </when> + <when value="no_restriction"/> + </conditional> + <conditional name="tabular_annotation"> + <param name="load_annotation" type="select" label="Use pixel annotation from tabular file to have updated annotation information in case empty spectra will be removed"> + <option value="no_annotation" selected="True">use no annotation</option> + <option value="yes_annotation">use pixel annotation from a tabular file</option> + </param> + <when value="yes_annotation"> + <param name="annotation_file" type="data" format="tabular" label="Use annotations from tabular file" + help="Tabular file with three columns: x values, y values and pixel annotations"/> + <param name="column_x" data_ref="annotation_file" label="Column with x values" type="data_column"/> + <param name="column_y" data_ref="annotation_file" label="Column with y values" type="data_column"/> + <param name="column_names" data_ref="annotation_file" label="Column with pixel annotations" type="data_column"/> + <param name="tabular_header" type="boolean" label="Tabular file contains a header line" truevalue="TRUE" falsevalue="FALSE"/> + </when> + <when value="no_annotation"/> + </conditional> + <repeat name="methods" title="Method" min="1"> + <conditional name="methods_conditional"> + <param name="method" type="select" label="Select the method you want to apply"> + <option value="Transformation" selected="True">Transformation</option> + <option value="Smoothing">Smoothing</option> + <option value="Baseline">Baseline removal</option> + <option value="Calibrate">Calibrate</option> + <option value="Align">Align Spectra (warping/phase correction)</option> + <validator type="empty_field" /> + </param> + <when value="Transformation"> + <param name="transform_method" type="select" label="Select your transfprormation method"> + <option value="sqrt" selected="True">sqrt</option> + <option value="log">log</option> + <option value="log2">log2</option> + <option value="log10">log10</option> + <validator type="empty_field" /> + </param> + </when> + <when value="Smoothing"> + <conditional name="methods_for_smoothing"> + <param name="smooth_method" type="select" label="This method smoothes the intensity values of a MassSpectrum object"> + <option value="SavitzkyGolay" selected="True">SavitzkyGolay</option> + <option value="MovingAverage">MovingAverage</option> + </param> + <when value="SavitzkyGolay"> + <param name="polynomial" value="3" type="text" label="PolynomialOrder argument to control the order of the filter"/> + </when> + <when value="MovingAverage"> + <param name="weighted" type="boolean" label="Weighted average" help = "indicates if the average should be equal weight or if it should have weights depending on the distance from the center as calculated as 1/2^abs(-halfWindowSize:halfWindowSize) with the sum of all weigths normalized to 1" truevalue="TRUE" falsevalue="FALSE"/> + </when> + </conditional> + <param name="halfWindowSize" type="integer" value="10" + label="Half window size" + help="The resulting window reaches from + mass[currentIndex-halfWindowSize] to mass[currentIndex+halfWindowSize] + (window size is 2*halfWindowSize+1). + The best size differs depending on the selected smoothing method."/> + </when> + <when value="Baseline"> + <param name="baseline_method" type="select" label="Baseline removal method"> + <option value="SNIP" selected="True">SNIP</option> + <option value="TopHat">TopHat</option> + <option value="ConvexHull">ConvexHull</option> + <option value="median">median</option> + <validator type="empty_field" /> + </param> + <param name="iterations" type="integer" value="100" + label="Number of iterations" + help=""/> + </when> + <when value="Calibrate"> + <param name="calibrate_method" type="select" label="Calibration method"> + <option value="TIC" selected="True">TIC</option> + <option value="PQN">PQN</option> + <option value="median">median</option> + <validator type="empty_field" /> + </param> + <param name="mass_start" type="integer" value="0" + label="Start of m/z range, has to be inside m/z range" + help="Scaling factor is calculated on the mass range and applied to the whole spectrum"/> + <param name="mass_end" type="integer" value="0" + label="End of m/z range, has to be inside m/z range" + help="The Start and End value needs to be different from 0 to be taken into account and."/> + </when> + <when value="Align"> + <param name="warping_method" type="select" label="Warping methods"> + <option value="lowess" selected="True">Lowess</option> + <option value="linear">Linear</option> + <option value="quadratic">Quadratic</option> + <option value="cubic">Cubic</option> + </param> + + <param name="tolerance" type="float" value="0.002" + label="Tolerance" + help="Double, maximal relative deviation of a peak position (m/z) to be considered as identical" /> + + <param name="halfWindowSize" type="integer" value="20" + label="Half window size" + help="The resulting window reaches from + mass[currentIndex-halfWindowSize] to mass[currentIndex+halfWindowSize] + (window size is 2*halfWindowSize+1). + The best size differs depending on the selected smoothing method."/> + + <param name="snr" type="integer" value="2" + label="Signal-to-noise-ratio" + help=""/> + + <conditional name="reference_for_alignment"> + <param name="align_ref" type="select" label="Reference to which the samples should be aligned" help="Use internal calibrants to perform m/z calibration"> + <option value="no_reference" selected="True">no reference</option> + <option value="yes_reference">reference from tabular file</option> + </param> + <when value="no_reference"/> + <when value="yes_reference"> + <param name="reference_file" type="data" format="tabular" + label="Tabular file with m/z of internal calibrants (MassPeaks) which should be used for spectra alignment" + help="calibration of m/z values to internal calibrants, at least 2 m/z per spectrum are needed"/> + <param name="allow_nomatch" type="boolean" label="Don't throw an error when less than 2 reference m/z were found in a spectrum" truevalue="TRUE" falsevalue="FALSE"/> + <param name="empty_nomatch" type="boolean" label="logical, if TRUE the intensity values of MassSpectrum or MassPeaks objects with missing (NA) warping functions are set to zero" truevalue="TRUE" falsevalue="FALSE"/> + <param name="remove_empty" type="boolean" label="Should empty spectra be removed" truevalue="TRUE" falsevalue="FALSE"/> + </when> + </conditional> + </when> + </conditional> + </repeat> + <param name="export_processed" type="boolean" label="Export file as processed imzML" help="otherwise continuous imzML will be exported" truevalue="TRUE" falsevalue="FALSE"/> + </inputs> + <outputs> + <data format="imzml" name="outfile_imzml" label="$infile.display_name processed" /> + <data format="pdf" name="plots" from_work_dir="prepro_qc_plot.pdf" label="$infile.display_name preprocessed QC"/> + <data format="tabular" name="annotation_output" label="$infile.display_name annotations"> + <filter>tabular_annotation["load_annotation"] == 'yes_annotation'</filter> + </data> + </outputs> + <tests> + <test> + <param name="infile" value="" ftype="imzml"> + <composite_data value="Example_Continuous.imzML"/> + <composite_data value="Example_Continuous.ibd"/> + </param> + <conditional name="restriction_conditional"> + <param name="restriction" value="restrict"/> + <param name="coordinates_file" value="restricted_pixels.tabular"/> + </conditional> + <conditional name="methods_conditional"> + <param name="method" value="Transformation"/> + <param name="transform_method" value="log2"/> + <param name="method" value="Smoothing"/> + <param name="smooth_method" value="SavitzkyGolay"/> + <param name="method" value="Basline"/> + <param name="baseline_method" value ="TopHat"/> + </conditional> + <output name="outfile_imzml" file="outfile1.imzML" compare="sim_size"/> + <output name="outfile_imzml" file="outfile1.ibd" compare="sim_size"/> + <output name="plots" file="Preprocessing1_QC.pdf" compare="sim_size"/> + </test> + <test> + <param name="infile" value="msidata_1.RData" ftype="rdata"/> + <conditional name="methods_conditional"> + <param name="method" value="Calibrate"/> + <param name="calibrate_method" value="PQN"/> + </conditional> + <output name="outfile_imzml" file="outfile2.imzML" compare="sim_size"/> + <output name="outfile_imzml" file="outfile2.ibd" compare="sim_size"/> + <output name="plots" file="Preprocessing2_QC.pdf" compare="sim_size"/> + </test> + <test> + <param name="infile" value="" ftype="imzml"> + <composite_data value="Example_Continuous.imzML"/> + <composite_data value="Example_Continuous.ibd"/> + </param> + <conditional name="tabular_annotation"> + <param name="load_annotation" value="yes_annotation"/> + <param name="annotation_file" value="pixel_annotations.tabular"/> + <param name="column_x" value="1"/> + <param name="column_y" value="2"/> + <param name="column_names" value="3"/> + <param name="tabular_header" value="TRUE"/> + </conditional> + <conditional name="methods_conditional"> + <param name="method" value="Align"/> + <param name="warping_method" value="linear"/> + <param name="halfWindowSize" value="1"/> + <conditional name="reference_for_alignment"> + <param name="align_ref" value="yes_reference"/> + <param name="reference_file" value="align_reference_test2.tabular" ftype="tabular"/> + <param name="allow_nomatch" value="TRUE"/> + <param name="remove_empty" value="TRUE"/> + <param name="empty_nomatch" value="TRUE"/> + </conditional> + </conditional> + <output name="outfile_imzml" file="outfile3.imzML" compare="sim_size"/> + <output name="outfile_imzml" file="outfile3.ibd" compare="sim_size"/> + <output name="plots" file="Preprocessing3_QC.pdf" compare="sim_size"/> + <output name="annotation_output" file="annotations_output3.tabular"/> + </test> + </tests> + <help><![CDATA[ + +MALDIquant_ provides a complete analysis pipeline for MALDI-TOF and other mass spectrometry data. So far we have only implemented the functionalities for mass spectrometry imaging data. + +Input data: + +- MSI data as imzML file (upload via the "composite" function) `Introduction to the imzml format <https://ms-imaging.org/wp/imzml/>`_ +- optinal tabular file with pixel coordinates to restrict reading of imzML file to coordinates of interest + +Options: + +- Transformation: transformation of intensities with log, log2, log10 and squareroot +- Smoothing: Smoothing of the peaks reduces noise and improves peak detection. Available smoothing methods are SavitzkyGolay and Moving Average +- Baseline reduction: Baseline reduction removes background intensity generated by chemical noise (common in MALDI datasets). Available methods are SNIP, TopHat,ConvexHull and median. +- Intensity calibration (normalization): Normalization of intensities to Total Ion Current (TIC), median spectrum, Probabilistic Quotient Normalization (PQN) +- Spectra alignment (warping):alignment for (re)calibration of m/z values + + +Output: + +- imzML file (imzML format can be continuous or processed) +- pdf with average mass spectra after each preprocessing step + +.. _MALDIquant: http://strimmerlab.org/software/maldiquant/ + + ]]> + </help> + <expand macro="citation"/> +</tool>