<tool id="abims_xcms_group" name="xcms.group" version="2.0.2">

    <description>Group peaks together across samples using overlapping m/z bins and calculation of smoothed peak distributions in chromatographic time.</description>

    <requirements>
        <requirement type="package" version="3.1.2">R</requirement>
        <requirement type="binary">Rscript</requirement>
        <requirement type="package" version="1.44.0">xcms</requirement>
        <requirement type="package" version="2.1">xcms_w4m_script</requirement>
    </requirements>

    <stdio>
        <exit_code range="1:" level="fatal" />
    </stdio>

    <command>
        xcms.r 
        xfunction group image $image method $methods.method sleep 0.001 
        #if $methods.method == "density":
            ## minsamp $methods.minsamp 
            minfrac $methods.minfrac 
            bw $methods.bw 
            mzwid $methods.mzwid
        #if $methods.density_options.option == "show":
            max $methods.density_options.max
        #end if 
        #elif $methods.method == "mzClust":
            mzppm $methods.mzppm
            mzabs $methods.mzabs
            minfrac $methods.minfrac
            ## minsamp $methods.minsamp
        #else:
            mzVsRTbalance $methods.mzVsRTbalance
            mzCheck $methods.mzCheck
            rtCheck $methods.rtCheck
            kNN $methods.kNN
        #end if
        &amp;&amp; (
        mv group.RData $xsetRData;
        mv Rplots.pdf $rplotsPdf
        );
        cat xset.log
   </command> 

    <inputs>
        <param name="image" type="data" format="rdata.xcms.raw,rdata.xcms.group,rdata.xcms.retcor,rdata" label="xset RData file" help="output file from another function xcms (xcmsSet, retcor etc.)" />
        <conditional name="methods">
            <param name="method" type="select" label="Method to use for grouping" help="[method] See the help section below">
                <option value="density" selected="true">density</option>
                <option value="mzClust" >mzClust</option>
                <option value="nearest" >nearest</option>
            </param>
            <when value="density">
                <param name="bw" type="integer" value="30" label="Bandwidth" help="[bw] bandwidth (standard deviation or half width at half maximum) of gaussian smoothing kernel to apply to the peak density chromatogram" />
                <param name="minfrac" type="float" value="0.5" label="Minimum fraction of samples necessary" help="[minfrac] in at least one of the sample groups for it to be a valid group" />
                <param name="mzwid" type="float" value="0.25" label="Width of overlapping m/z slices" help="[mzwid] to use for creating peak density chromatograms and grouping peaks across samples " />
<!--
                <param name="minsamp" type="hidden" value="1" label="minsamp" help="minimum number of samples necessary in at least one of the sample groups for it to be a valid group " /> 
-->
            <conditional name="density_options">
            <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="max" type="integer" value="5" label="Maximum number of groups to identify in a single m/z slice" help="[max]" />
            </when>
            <when value="hide">
            </when>
        </conditional>

            </when>
            <when value="mzClust">
                <param name="mzppm" type="integer" value="20 " label="Relative error used for clustering/grouping in ppm" help="[mzppm]" />
                <param name="mzabs" type="float" value="0" label="Absolute error used for clustering/grouping" help="[mzabs]" />
                <param name="minfrac" type="float" value="0" label="Minimum fraction of each class in one bin" help="[minfrac] minimum fraction of samples necessary in at least one of the sample groups for it to be a valid group" />
<!--
                <param name="minsamp" type="hidden" value="1" label="minsamp" help="minimum number of samples necessary in at least one of the sample groups for it to be a valid group " />   
-->
            </when>
            <when value="nearest">
                <param name="mzVsRTbalance" type="integer" value="10 " label="Multiplicator for mz value before calculating the (euclidean) distance between two peaks." help="[mzVsRTbalance]" />
                <param name="mzCheck" type="float" value="0.2" label="Maximum tolerated distance for mz" help="[mzCheck]" />
                <param name="rtCheck" type="integer" value="15" label="Maximum tolerated distance for RT" help="[rtCheck]" />
                <param name="kNN" type="integer" value="10" label="Number of nearest Neighbours to check" help="[kNN]" />    
            </when>
        </conditional>
<!--
        <param name="sleepy" type="float" value="0.001" label="sleep" help="seconds to pause between plotting successive steps of the peak grouping algorithm. peaks are plotted as points showing relative intensity. identified groups are flanked by dotted vertical lines"> 
            <validator type="in_range" message="Must be more than 0" min="0.001" max="inf"/>
        </param>
-->

    </inputs>

    <outputs>
        <data name="xsetRData" format="rdata.xcms.group" label="${image.name[:-6]}.group.RData"/>
        <data name="rplotsPdf" format="pdf" label="${image.name[:-6]}.group.Rplots.pdf"/>
    </outputs>

    <tests>
        <test>
            <param name="image" value="xset.RData"/>
            <param name="methods.method" value="density"/>
            <param name="methods.bw" value="5"/>
            <param name="methods.minfrac" value="0.3"/>
            <param name="methods.mzwid" value="0.01"/>
            <param name="methods.density_options.option" value="show"/>
            <param name="methods.density_options.max" value="50"/>
            <output name="xsetRData" file="xset.group.RData" />
            <output name="rplotsPdf" file="xset.group.Rplots.pdf" />
         
        </test>
    </tests>

    <help>
        
.. class:: infomark

**Authors**  Colin A. Smith csmith@scripps.edu, Ralf Tautenhahn rtautenh@gmail.com, Steffen Neumann sneumann@ipb-halle.de, Paul Benton hpaul.benton08@imperial.ac.uk and Christopher Conley cjconley@ucdavis.edu 

**Galaxy integration** ABiMS TEAM - UPMC/CNRS - Station biologique de Roscoff and Yann Guitton yann.guitton@univ-nantes.fr - part of Workflow4Metabolomics.org [W4M]

 | Contact support@workflow4metabolomics.org for any questions or concerns about the Galaxy implementation of this tool.



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

==========
Xcms.Group
==========

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

After peak identification with xcmsSet, this tool groups the peaks which represent the same analyte across samples using overlapping m/z bins and calculation of smoothed peak distributions in chromatographic time. Allows rejection of features, which are only partially detected within the replicates of a sample class.



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

**Upstream tools**

========================= ================= =================== ==========
Name                      output file       format              parameter
========================= ================= =================== ==========
xcms.xcmsSet              xset.RData        rdata.xcms.raw      RData file
------------------------- ----------------- ------------------- ----------
xcms.retcor               xset.RData        rdata.xcms.retcor   RData file
========================= ================= =================== ==========


**Downstream tools**

+---------------------------+--------------------------------------+
| Name                      | Output file     | Format             |
+===========================+=================+====================+
|xcms.retcor                | xset.RData      | rdata.xcms.group   |
+---------------------------+--------------------------------------+
|xcms.fillPeaks             | xset.RData      | rdata.xcms.group   |
+---------------------------+--------------------------------------+

The output file is an xcmsSet.RData file. You can continue your analysis using it in **xcms.retcor** tool as an next step and then **xcms.fillPeaks**.

**General schema of the metabolomic workflow**

.. image:: xcms_group_workflow.png


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

+---------------------------+-----------------------+
| Parameter : num + label   |   Format              |
+===========================+=======================+
| 1 : RData file            |   rdata.xcms.group    |
+---------------------------+-----------------------+


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

Method to use for grouping
--------------------------

**mzClust**

    | Runs high resolution alignment on single spectra samples stored in the RData file generated by the **xcmsSet tool**.

**density**

    | Groups peaks together across samples using overlapping m/z bins and calculation of smoothed peak distributions in chromatographic time.

**nearest**

    | Groups peaks together across samples by creating a master peak list and assigning corresponding peaks from all samples. It is inspired by the alignment algorithm of mzMine.


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

xset.group.Rplots.pdf

xset.group.RData: rdata.xcms.group format

    | Rdata file that will be necessary in the third and fourth step of the workflow (xcms.retcor and xcms.fillpeaks).
    

------

.. class:: infomark 

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

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


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

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

    | RData file -> **xset.RData**

Parameters
----------

    | Method -> **density**
    | bw     -> **5**
    | minfrac -> **0.3**
    | mzwid    -> **0.01**
    | Advanced options: **show**
    | max -> **50**


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

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

    | **2) Example of an xset.group.Rplots pdf file**

.. image:: xcms_group.png
        :width: 700



    </help>


    <citations>
        <citation type="doi">10.1021/ac051437y</citation>
        <citation type="doi">10.1093/bioinformatics/btu813</citation>
    </citations>


</tool>