<tool id="abims_xcms_xcmsSet" name="xcms.xcmsSet" version="2.1.1">
    <description>Filtration and Peak Identification using xcmsSet function from xcms R package to preprocess LC/MS data for relative quantification and statistical analysis </description>

    <macros>
        <import>macros.xml</import>
    </macros>

    <expand macro="requirements"/>
    <expand macro="stdio"/>

    <command><![CDATA[
        @COMMAND_XCMS_SCRIPT@

        #if $input.is_of_type("mzxml") or $input.is_of_type("mzml") or $input.is_of_type("mzdata") or $input.is_of_type("netcdf"):
            singlefile_galaxyPath '$input' singlefile_sampleName '$input.name'
        #else
            zipfile '$input'
        #end if

        xfunction xcmsSet

        xsetRdataOutput '$xsetRData'
        sampleMetadataOutput '$sampleMetadata'
        ticspdf '$ticsRawPdf'
        bicspdf '$bpcsRawPdf'


        #if $options_scanrange.option == "show":
            scanrange "c($options_scanrange.scanrange)"
        #end if

        ## profmethod $profmethod
        nSlaves \${GALAXY_SLOTS:-1} method $methods.method
        #if $methods.method == "centWave":
            ppm $methods.ppm
            peakwidth "c($methods.peakwidth)"
            #if $methods.options_c.option == "show":
                mzdiff $methods.options_c.mzdiff
                snthresh $methods.options_c.snthresh
                integrate $methods.options_c.integrate
                noise $methods.options_c.noise
                prefilter "c($methods.options_c.prefilter)"
            #end if
        #elif $methods.method == "matchedFilter":
            step $methods.step
            fwhm $methods.fwhm
            #if $methods.options_m.option == "show":
                ## sigma "$methods.options_m.sigma"
                max $methods.options_m.max
                snthresh $methods.options_m.snthresh
                steps $methods.options_m.steps
                mzdiff $methods.options_m.mzdiff
            #end if
        #elif $methods.method == "MSW":
            snthr $methods.snthr
            nearbyPeak $methods.nearbyPeak
            winSize.noise $methods.winSize_noise
            amp.Th $methods.amp_Th
            scales "c($methods.scales)"
            SNR.method "$methods.SNR_method"
        #end if
        @COMMAND_LOG_EXIT@
    ]]></command>

    <inputs>

        <param name="input" type="data" format="mzxml,mzml,mzdata,netcdf,no_unzip.zip,zip" label="File(s) from your history containing your chromatograms" help="Single file mode for the format: mzxml, mzml, mzdata and netcdf. Zip file mode for the format: no_unzip.zip, zip. See the help section below." />

        <conditional name="options_scanrange">
            <param name="option" type="select" label="Scan range option " >
                <option value="show">show</option>
                <option value="hide" selected="true">hide</option>
            </param>
            <when value="show">
                <param name="scanrange" type="text" value="" label="scanrange" help="scan range to process, for example (16,365)" >
                    <validator type="empty_field"/>
                </param>
            </when>
            <when value="hide">
            </when>
        </conditional>


<!--
        <param name="profmethod" type="select" label="Method to use for profile generation (profmethod)" >
            <option value="bin" selected="true">bin</option>
            <option value="binlin">binlin</option>
            <option value="binlinbase">binlinbase</option>
            <option value="intlin">intlin</option>
        </param>
-->
        <conditional name="methods">
            <param name="method" type="select" label="Extraction method for peaks detection" help="[method] See the help section below">
                <option value="centWave" >centWave</option>
                <option value="matchedFilter" selected="true">matchedFilter</option>
                <option value="MSW">MSW</option>
            </param>

            <!-- centWave Filter options -->
            <when value="centWave">
                <param name="ppm" type="integer" value="25" label="Max tolerated ppm m/z deviation in consecutive scans in ppm" help="[ppm]" />
                <param name="peakwidth" type="text" value="20,50" label="Min,Max peak width in seconds" help="[peakwidth]" />

                <conditional name="options_c">
                    <param name="option" type="select" label="Advanced options" >
                        <option value="show">show</option>
                        <option value="hide" selected="true">hide</option>
                    </param>
                    <when value="show">
                        <param name="snthresh" type="integer" value="10" label="Signal/Noise threshold" help="[snthresh] Signal to noise ratio cutoff" />
                        <param name="mzdiff" type="float" value="-0.001" label="Minimum difference in m/z for peaks with overlapping retention times" help="[mzdiff] Can be negative to allow overlap" />
                        <param name="integrate" type="select" label="peak limits method" help="[integrate]" >
                            <option value="1">peak limits based on smoothed 2nd derivative (less precise)</option>
                            <option value="2">peak limits based on real data (more sensitive to noise)</option>
                        </param>
                        <param name="prefilter" type="text" value="3,100" label="Prefilter step for the first phase" help="[prefilter] Separate by coma k,I. Mass traces are only retained if they contain at least ‘k’ peaks with intensity >= ‘I’"/>
                        <param name="noise" type="integer" value="0" label="Noise filter" help="[noise] optional argument which is useful for data that was centroided without any intensity threshold, centroids with intensity smaller than ‘noise’ are omitted from ROI detection"/>
                    </when>
                    <when value="hide">
                    </when>
                </conditional>
            </when>

            <!-- matched Filter options -->
            <when value="matchedFilter">
                <param name="step" type="float" value="0.1" label="Step size to use for profile generation" help="[step] The peak detection algorithm creates extracted ion base peak chromatograms (EIBPC) on a fixed step size" />
                <param name="fwhm" type="integer" value="30" label="Full width at half maximum of matched filtration gaussian model peak" help="[fwhm] Only used to calculate the actual sigma" />
                <conditional name="options_m">
                    <param name="option" type="select" label="Advanced options" >
                        <option value="show">show</option>
                        <option value="hide" selected="true">hide</option>
                    </param>
                    <when value="show">
                        <!--<param name="sigma" type="float" value="12.739935451" label="Standard deviation (width) of matched filtration model peak" help="[sigma] By default: fwhm/2.3548" />-->
                        <param name="max" type="integer" value="5" label="Maximum number of peaks per extracted ion chromatogram" help="[max]" />
                        <param name="snthresh" type="integer" value="10" label="Signal to noise ratio cutoff" help="[snthresh]" />
                        <param name="steps" type="integer" value="2" label="Number of steps to merge prior to filtration" help="[steps] The peak identification algorithm combines a given number of EIBPCs prior to filtration and peak detection, as defined by the steps argument" />
                        <param name="mzdiff" type="float" value="0.6" label="Minimum difference in m/z for peaks with overlapping Retention Times" help="[mzdiff] By default: 0.8-step*steps " />
                    </when>
                    <when value="hide">
                    </when>
                </conditional>
            </when>

            <!-- MSW Filter options -->
            <when value="MSW">
                <param name="nearbyPeak" type="select" label="Determine whether to include the nearby small peaks of major peaks" help="[nearbyPeak]" >
                    <option value="TRUE">TRUE</option>
                    <option value="FALSE">FALSE</option>
                </param>
                <param name="winSize_noise" type="integer" value="500" label="The local window size to estimate the noise level" help="[winSize.noise]" />
                <param name="snthr" type="integer" value="3" label="SNR (Signal to Noise Ratio) threshold" help="[snthr]" />
                <param name="amp_Th" type="float" value="0.002" label="Minimum required relative amplitude of the peak" help="[amp.Th] Ratio to the maximum of CWT coefficients" />
                <param name="scales" type="text" value="seq(1,22,3)" label="Scales for the Continuous Wavelet Transform (CWT)" help="[scales] Scales are linked to the width of the peaks that are to be detected. Tape as indicaded seq('n,n,n') or c(n,n) : seq(from, to, by steps), c - linear vector " />
                <param name="SNR_method" type="text" value="data.mean" label="SNR (Signal to Noise Ratio) method" help="[SNR.method] Method to estimate noise level. Currently, only 95 percentage quantile is supported." />
            </when>
        </conditional>
    </inputs>

    <outputs>
        <data name="xsetRData" format="rdata.xcms.raw" label="${input.name.rsplit('.',1)[0]}.xset.RData" />
        <data name="sampleMetadata" format="tabular" label="${input.name.rsplit('.',1)[0]}.sampleMetadata.tsv">
            <filter>input.extension not in ["mzxml","mzml","mzdata","netcdf"]</filter>
        </data>
        <data name="ticsRawPdf"   format="pdf" label="${input.name.rsplit('.',1)[0]}.xset.TICs_raw.pdf" />
        <data name="bpcsRawPdf"   format="pdf" label="${input.name.rsplit('.',1)[0]}.xset.BPCs_raw.pdf" />
        <data name="log" format="txt" label="${input.name.rsplit('.',1)[0]}.xset.log.txt" />
    </outputs>

    <tests>
        <!--<test>
            <param name="input" value="sacuri_dir_root.zip"  ftype="zip" />
            <param name="methods|method" value="matchedFilter" />
            <param name="methods|step" value="0.01" />
            <param name="methods|fwhm" value="4" />
            <param name="methods|options_m|option" value="show" />
            <param name="methods|options_m|max" value="50" />
            <param name="methods|options_m|snthresh" value="1" />
            <param name="methods|options_m|steps" value="2" />
            <output name="log">
                <assert_contents>
                    <has_text text="object with 4 samples" />
                    <has_text text="Time range: 0.7-1139.7 seconds (0-19 minutes)" />
                    <has_text text="Mass range: 50.0021-999.9863 m/z" />
                    <has_text text="Peaks: 59359 (about 14840 per sample)" />
                    <has_text text="Peak Groups: 0" />
                    <has_text text="Sample classes: bio, blank" />
                </assert_contents>
            </output>
        </test>
        <test>
            <param name="input" value="sacuri_current_root.zip"  ftype="zip" />
            <param name="methods|method" value="centWave" />
            <param name="methods|ppm" value="25" />
            <param name="methods|peakwidth" value="20,50" />
            <output name="log">
                <assert_contents>
                    <has_text text="object with 4 samples" />
                    <has_text text="Time range: 3.5-1139.2 seconds (0.1-19 minutes)" />
                    <has_text text="Mass range: 57.9756-593.4086 m/z" />
                    <has_text text="Peaks: 1535 (about 384 per sample)" />
                    <has_text text="Peak Groups: 0" />
                    <has_text text="Sample classes: bio, blank" />
                </assert_contents>
            </output>
        </test>-->
        <test>
            <param name="input" value="faahKO_reduce.zip"  ftype="zip" />
            <param name="methods|method" value="centWave" />
            <param name="methods|ppm" value="25" />
            <param name="methods|peakwidth" value="20,50" />
            <output name="log">
                <assert_contents>
                    <has_text text="object with 4 samples" />
                    <has_text text="Time range: 2506.1-4477.9 seconds (41.8-74.6 minutes)" />
                    <has_text text="Mass range: 200.1-600 m/z" />
                    <has_text text="Peaks: 9251 (about 2313 per sample)" />
                    <has_text text="Peak Groups: 0" />
                    <has_text text="Sample classes: KO, WT" />
                </assert_contents>
            </output>
        </test>
        <!-- Passed but disable to save time for Travis" -->
        <!--<test>
            <param name="input" value="ko15.CDF"  ftype="netcdf" />
            <param name="methods|method" value="centWave" />
            <param name="methods|ppm" value="25" />
            <param name="methods|peakwidth" value="20,50" />
            <output name="log">
                <assert_contents>
                    <has_text text="object with 1 samples" />
                    <has_text text="Time range: 2506.1-4471.7 seconds (41.8-74.5 minutes)" />
                    <has_text text="Mass range: 200.2-600 m/z" />
                    <has_text text="Peaks: 2262 (about 2262 per sample)" />
                    <has_text text="Peak Groups: 0" />
                    <has_text text="Sample classes: ." />
                </assert_contents>
            </output>
        </test>
        <test>
            <param name="input" value="ko16.CDF"  ftype="netcdf" />
            <param name="methods|method" value="centWave" />
            <param name="methods|ppm" value="25" />
            <param name="methods|peakwidth" value="20,50" />
            <output name="log">
                <assert_contents>
                    <has_text text="object with 1 samples" />
                    <has_text text="Time range: 2521.7-4477.9 seconds (42-74.6 minutes)" />
                    <has_text text="Mass range: 200.1-600 m/z" />
                    <has_text text="Peaks: 2408 (about 2408 per sample)" />
                    <has_text text="Peak Groups: 0" />
                    <has_text text="Sample classes: ." />
                </assert_contents>
            </output>
        </test>
        <test>
            <param name="input" value="wt15.CDF"  ftype="netcdf" />
            <param name="methods|method" value="centWave" />
            <param name="methods|ppm" value="25" />
            <param name="methods|peakwidth" value="20,50" />
            <output name="log">
                <assert_contents>
                    <has_text text="object with 1 samples" />
                    <has_text text="Time range: 2517-4473.2 seconds (42-74.6 minutes)" />
                    <has_text text="Mass range: 200.2-599.8 m/z" />
                    <has_text text="Peaks: 2278 (about 2278 per sample)" />
                    <has_text text="Peak Groups: 0" />
                    <has_text text="Sample classes: ." />
                </assert_contents>
            </output>
        </test>
        <test>
            <param name="inputs|input" value="single_file" />
            <param name="inputs|single_file" value="wt16.CDF"  ftype="netcdf" />
            <param name="methods|method" value="centWave" />
            <param name="methods|ppm" value="25" />
            <param name="methods|peakwidth" value="20,50" />
            <output name="log">
                <assert_contents>
                    <has_text text="object with 1 samples" />
                    <has_text text="Time range: 2521.7-4468.5 seconds (42-74.5 minutes)" />
                    <has_text text="Mass range: 200.3-600 m/z" />
                    <has_text text="Peaks: 2303 (about 2303 per sample)" />
                    <has_text text="Peak Groups: 0" />
                    <has_text text="Sample classes: ." />
                </assert_contents>
            </output>
        </test>-->
        <test>
            <param name="input" value="HU_neg_017.mzXML"  ftype="mzxml" />
            <param name="methods|method" value="centWave" />
            <param name="methods|ppm" value="25" />
            <param name="methods|peakwidth" value="20,50" />
            <output name="log">
                <assert_contents>
                    <has_text text="object with 1 samples" />
                    <has_text text="Time range: 3.5-1139.1 seconds (0.1-19 minutes)" />
                    <has_text text="Mass range: 57.9756-556.8128 m/z" />
                    <has_text text="Peaks: 380 (about 380 per sample)" />
                    <has_text text="Peak Groups: 0" />
                    <has_text text="Sample classes: ." />
                </assert_contents>
            </output>
        </test>
        <test>
            <param name="input" value="MM14.mzML"  ftype="mzxml" />
            <param name="methods|method" value="centWave" />
            <param name="methods|ppm" value="56" />
            <param name="methods|peakwidth" value="5.275,13.5" />
            <output name="log">
                <assert_contents>
                    <has_text text="object with 1 samples" />
                    <has_text text="Time range: 271-307.3 seconds (4.5-5.1 minutes)" />
                    <has_text text="Mass range: 117.0357-936.7059 m/z" />
                    <has_text text="Peaks: 222 (about 222 per sample)" />
                    <has_text text="Peak Groups: 0" />
                    <has_text text="Sample classes: ." />
                </assert_contents>
            </output>
        </test>
    </tests>

    <help><![CDATA[

@HELP_AUTHORS@

============
Xcms.xcmsSet
============

-----------
Description
-----------

This tool is used for preprocessing analyte data from multiple LC/MS files (formats NetCDF, mzXML and mzData). It extracts ion from each sample independently and using a statistic model, peaks are filtered and integrated.
You can read a tutorial on how to perform xcms preprocessing which is available here_.

.. _here: http://web11.sb-roscoff.fr/download/w4m/howto/w4m_HowToPerformXcmsPreprocessing_v02.pdf


-----------------
Workflow position
-----------------

**Upstream tools**

========================= ================= ======= =========
Name                      output file       format  parameter
========================= ================= ======= =========
NA                        NA                zip     NA
========================= ================= ======= =========


**Downstream tools**

+---------------------------+--------------------+-----------------+
| Name                      | Output file        | Format          |
+===========================+====================+=================+
|xcms.group                 | xset.RData         | rdata.xcms.raw  |
+---------------------------+--------------------+-----------------+
|PCA ellipsoid by factors   | sampleMetadata.tsv | Tabular         |
+---------------------------+--------------------+-----------------+
|Anova                      | sampleMetadata.tsv | Tabular         |
+---------------------------+--------------------+-----------------+


**Example of a metabolomic workflow**

.. image:: xcms_xcmsset_workflow.png


------

.. class:: infomark

The output file is an xset.RData file. You can continue your analysis using it in **xcms.group** tool.

---------------------------------------------------



-----------
Input files
-----------

+---------------------------+----------------------------------+
| Parameter : num + label   |   Format                         |
+===========================+==================================+
| OR : Zip file             |   zip                            |
+---------------------------+----------------------------------+
| OR : Single file          |   mzXML, mzML, mzData, netCDF    |
+---------------------------+----------------------------------+

**Choose your inputs**

You have two methods for your inputs:

    | Single file (recommended): You can put a single file as input. That way, you will be able to launch several xcmsSet in parallel and use "xcms.xcmsSet Merger" before "xcms.group"
    | Zip file: You can put a zip file containing your inputs: myinputs.zip (containing all your conditions as sub-directories).

Zip file: Steps for creating the zip file
-----------------------------------------

**Step1: Creating your directory and hierarchize the subdirectories**


VERY IMPORTANT: If you zip your files under Windows, you must use the 7Zip software (http://www.7-zip.org/), otherwise your zip will not be well unzipped on the platform W4M (zip corrupted bug).

Your zip should contain all your conditions as sub-directories. For example, two conditions (mutant and wild):
arabidopsis/wild/01.raw
arabidopsis/mutant/01.raw

**Step2: Creating a zip file**

Create your zip file (e.g.: arabidopsis.zip).

**Step 3 : Uploading it to our Galaxy server**

If your zip file is less than 2Gb, you get use the Get Data tool to upload it.

Otherwise if your zip file is larger than 2Gb, please refer to the HOWTO on workflow4metabolomics.org (http://application.sb-roscoff.fr/download/w4m/howto/galaxy_upload_up_2Go.pdf).

For more informations, don't hesitate to send us an email at supportATworkflow4metabolomics.org).

Advices for converting your files for the XCMS input
----------------------------------------------------

We recommend you to convert your raw files to **mzXML** in centroid mode (smaller files) and the files will be compatible with the xmcs centWave method.

**We recommend you the following parameters:**

Use Filtering: **True**

Use Peak Picking: **True**

Peak Peaking -Apply to MS Levels: **All Levels (1-)** : Centroid Mode

Use zlib: **64**

Binary Encoding: **64**

m/z Encoding: **64**

Intensity Encoding: **64**


----------
Parameters
----------

Extraction method for peaks detection
-------------------------------------

**Matched Filter**

    | One parameter to consider is the Gaussian model peak width used for matched filtration,an integral part of the peak detection algorithm.
    | For a discussion of how model peak width affects the signal to noise ratio, see Danielsson et al. (2002).


**cent Wave**

    | This algorithm is most suitable for high resolution LC/{TOF,OrbiTrap,FTICR}-MS data in centroid mode.
    | Due to the fact that peak centroids are used, a binning step is not necessary.
    | The method is capable of detecting close-by-peaks and also overlapping peaks. Some efforts are made to detect the exact peak boundaries to get precise peak integrals.

**MSW**

    | Wavelet based, used for direct infusion data. Continuous wavelet transform (CWT) can be used to locate chromatographic peaks on different scales.
    | If you wish to have more details about the other parameters, you can read the following documents:
    | -Example of preprocessing data with XCMS : http://www.bioconductor.org/packages/2.12/bioc/vignettes/xcms/inst/doc/xcmsPreprocess.pdf
    | -Details and explanations for all the parameters of XCMS package: http://www.bioconductor.org/packages/release/bioc/manuals/xcms/man/xcms.pdf


------------
Output files
------------

xset.TICs_raw.pdf

    | "Total Ion Chromatograms" graph in pdf format.

xset.BPCs_raw.pdf

    | "Base Peak Chromatograms" graph in pdf format with each class samples opposed.

sampleMetadata.tsv

    | Tabular file that contains for each sample, it's associated class and polarity (positive,negative and mixed).
    | This file is necessary in the Anova and PCA step of the workflow.

xset.RData: rdata.xcms.raw format

    | Rdata file that is necessary in the second step of the workflow "xcms.group".

------

.. class:: infomark

The output file is an xset.RData file. You can continue your analysis using it in **xcms.group** tool.

---------------------------------------------------

---------------
Working example
---------------

Input files
-----------

    | zip_file -> **sacuri.zip**

Parameters
----------

    | Method -> **matchedFilter**
    | step   -> **0.01**
    | fwhm   -> **4**
    | Advanced option -> **show**
    | max: -> **50**
    | snthresh -> **1**
    | steps -> **2**


Output files
------------

    | **1) xset.RData: RData file**

    | **2) Example of a sampleMetadata.tsv  :**


+---------------------------+------------+---------+
| sampleMetadata            |   class    | polarity|
+===========================+============+=========+
|HU_neg_017                 |   bio      |negative |
+---------------------------+------------+---------+
|HU_neg_028                 |   bio      |negative |
+---------------------------+------------+---------+
|HU_neg_034                 |   bio      |negative |
+---------------------------+------------+---------+
|Blanc04                    |   blank    |negative |
+---------------------------+------------+---------+
|Blanc06                    |   blank    |negative |
+---------------------------+------------+---------+
|Blanc09                    |   blank    |negative |
+---------------------------+------------+---------+



    | **3) Example of xset.TICs_raw.pdf (Total Ion Chromatograms) :**

.. image:: xcms_tics.png


---------------------------------------------------

Changelog/News
--------------

**Version 2.1.1 - 29/11/2017**

- BUGFIX: To avoid issues with accented letter in the parentFile tag of the mzXML files, we changed a hidden mechanim to LC_ALL=C

**Version 2.1.0 - 22/02/2017**

- NEW: The W4M tools will be able now to take as input a single file. It will allow to submit in parallel several files and merge them afterward using "xcms.xcmsSet Merger" before "xcms.group".

- BUGFIX: the default value of "matchedFilter" -> "Step size to use for profile generation" which was of 0.01 have been changed to fix with the XMCS default values to 0.1

**Version 2.0.11 - 22/12/2016**

- BUGFIX: propose scanrange for all methods

**Version 2.0.10 - 22/12/2016**

- BUGFIX: when having only one group (i.e. one folder of raw data) the BPC and TIC pdf files do not contain any graph

**Version 2.0.9 - 06/07/2016**

- UPGRADE: upgrade the xcms version from 1.44.0 to 1.46.0

**Version 2.0.8 - 06/04/2016**

- TEST: refactoring to pass planemo test using conda dependencies


**Version 2.0.7 - 10/02/2016**

- BUGFIX: better management of errors. Datasets remained green although the process failed

- BUGFIX/IMPROVEMENT: New checking steps around the imported data in order to raise explicte error message before or after launch XCMS: checking of bad characters in the filenames, checking of the XML integrity and checking of duplicates which can appear in the sample names during the XCMS process because of bad characters

- BUGFIX/IMPROVEMENT: New step to check and delete bad characters in the XML: accented characters in the storage path of the mass spectrometer

- UPDATE: refactoring of internal management of inputs/outputs

- TEST: refactoring to feed the new report tool


**Version 2.0.2 - 18/01/2016**

- BUGFIX: Some zip files were tag as "corrupt" by R. We have changed the extraction mode to deal with thoses cases.


**Version 2.0.2 - 09/10/2015**

- BUGFIX: Some users reported a bug in xcms.xcmsSet. The preprocessing stops itself and doesn't import the whole dataset contained in the zip file without warning. But meanwhile, please check your samplemetadata dataset and the number of rows.


**Version 2.0.2 - 02/06/2015**

- NEW: The W4M workflows will now take as input a zip file to ease the transfer and to improve dataset exchange between tools and users. (See How_to_upload). The previous "Library directory name" is still available but we invite user to switch on the new zip system as soon as possible.

- IMPROVEMENT: new datatype/dataset formats (rdata.xcms.raw, rdata.xcms.group, rdata.xcms.retcor ...) will facilitate the sequence of tools and so avoid incompatibility errors.

- IMPROVEMENT: parameter labels have changed to facilitate their reading.

    ]]></help>

    <expand macro="citation" />
</tool>