<tool id="mutational_patterns" name="Analyse Mutational Patterns/Signatures" version="@TOOL_VERSION@+galaxy@VERSION_SUFFIX@" profile="@PROFILE@">
    <description>from genomic variations in vcf files</description>
    <macros>
        <token name="@TOOL_VERSION@">3.12.0</token>
        <token name="@VERSION_SUFFIX@">1</token>
        <token name="@PROFILE@">23.0</token>
    </macros>
    <requirements>
        <requirement type="package" version="@TOOL_VERSION@">bioconductor-mutationalpatterns</requirement>
        <requirement type="package" version="3.4.4">r-ggplot2</requirement>
        <requirement type="package" version="1.7.4">r-optparse</requirement>
        <requirement type="package" version="0.2.21">r-rjson</requirement>
        <requirement type="package" version="0.21.0">r-nmf</requirement>
        <requirement type="package" version="2.3">r-gridextra</requirement>
        <requirement type="package" version="1.4.3">bioconductor-bsgenome.hsapiens.ucsc.hg19</requirement>
        <requirement type="package" version="1.4.5">bioconductor-bsgenome.hsapiens.ucsc.hg38</requirement>
<!--
install more bioconda genomes
bioconductor-bsgenome.mmusculus.ucsc.mm9
bioconductor-bsgenome.mmusculus.ucsc.mm10
-->
    </requirements>
    <stdio>
        <exit_code range="1:" level="fatal" description="Tool exception" />
    </stdio>

    <command detect_errors="exit_code"><![CDATA[
#import json
#import os
Rscript $__tool_directory__/mutational_patterns.R
    --inputs
    #set $filename_to_element_identifiers = {}
    #for $sample in $vcfs:
        $filename_to_element_identifiers.__setitem__(str($sample),  $sample.element_identifier)
    #end for
    '#echo json.dumps(filename_to_element_identifiers)#'
    --genome '$genome'
    
    #if $set_levels.choices == 'yes':
      --levels '$set_levels.levels'
    #end if

    #if $set_spectrum.choices == 'yes':
        --output_spectrum '$spectrum'
    #end if
    
    #if $set_denovo.choices == 'yes':
        --nrun '$set_denovo.nrun'
        --rank '$set_denovo.rank'
        --newsignum '$set_denovo.newsignum'
        --output_denovo '$denovo'
        --sigmatrix '$sigmatrix'
        --cosmic_id_threshold '$set_denovo.cosmic_id_threshold'
    #end if

    #if $set_preset.choices == 'yes':

        #if $set_preset.set_signature_input.input_signature_choices == 'cosmic'
            --cosmic_version '$set_preset.set_signature_input.cosmic_version'
        #end if

        #if $set_preset.set_signature_input.input_signature_choices == 'user_signatures'
            --own_signatures '$set_preset.set_signature_input.own_matrix'
        #end if
       
        --signum '$set_preset.signum'
        --colors '$set_preset.colors'
        --display_signatures '$set_preset.display_signatures'
        --output_sigpattern '$sig_contrib'

        #if $set_preset.contrib_matrix_out == 'yes'
            --sig_contrib_matrix '$sig_contrib_matrix'
        #end if

    #end if
            
    #if $rdata_out
        --rdata '$rdata'
    #end if
    
    --tooldir '$__tool_directory__/'

]]></command>
    <inputs>
        <param name="vcfs" type="data_collection" format="vcf" label="VCF file(s) collection" multiple="true"/>
        <param name="genome" type="select" label="Reference Genome">
            <option value="BSgenome.Hsapiens.UCSC.hg19">BSgenome.Hsapiens.UCSC.hg19</option>
            <option value="BSgenome.Hsapiens.UCSC.hg38" selected="true">BSgenome.Hsapiens.UCSC.hg38</option>
        </param>
        <conditional name="set_levels">
            <param name="choices" type="select" label="samples have levels/labels for grouping them in the analysis" display="radio"
                   help="for instance, female/male or genotype-1/genotype-2/genotype-3, etc.">
                <option value="no" selected="true">No</option>
                <option value="yes">Yes</option>
            </param>
            <when value="yes">
                <param name="levels" type="data" format="tabular"
                       label="A two-column tab-separated file describing levels attributed to each sample name"
                       help="Tip: the sample name list in the vcf collection can be obtained using
                       the IUC Galaxy tool 'Extract element identifiers of a list collection' &lt;br&gt;
                       example:&lt;br&gt;
                       sample1 female&lt;br&gt;
                       sample2 female&lt;br&gt;
                       sample3 male&lt;br&gt;
                       sample4 female&lt;br&gt;
                       sample5 male&lt;br&gt;
                       sample5 male" />
            </when>
            <when value="no" />
        </conditional>
        <conditional name="set_spectrum">
            <param name="choices" type="select" label="Analyse Mutational Spectrum" display="radio">
                <option value="yes" selected="true">Yes</option>
                <option value="no">No</option>
            </param>
            <when value="yes" />
            <when value="no" />
        </conditional>

        <conditional name="set_denovo">
            <param name="choices" type="select" label="Extract de novo signatures with MutationalPatterns" display="radio">
                <option value="yes">Yes</option>
                <option value="no" selected="true">No</option>
            </param>
            <when value="yes">
                <param name="nrun" type="integer" value="10" min="10" max="200"
                       label="Number of cycles to find best fitting of signatures"
                       help="High values extend the computational time"/>
                <param name="rank" type="integer" value="4" min="3" max="30"
                       label="Number of ranks to be displayed for control of optimal number of signature"
                       help="High values extend the computational time. In addition the number of ranks cannot be greater than the number of available samples in the study"/>
                <param name="newsignum" type="integer" value="4" min="2" max="30"
                       label="Number of de novo signatures to capture"
                       help="High values extend the computational time. Note also that you cannot extract more signature than the number of available samples in the study"/>
                <param name="cosmic_id_threshold" type="float" value="0.85" min="0" max="1"
                       label="Minimum cosine similarity to rename a signature"
                       help="If a signature has a cosine similarity higher than this value with a cosmic v3.2 signature, it will renamed accordingly, with a -like suffix"/>
            </when>
            <when value="no" />
        </conditional>

        <conditional name="set_preset">
            <param name="choices" type="select" label="Decompose with preset signatures" display="radio">
                <option value="yes" selected="true">Yes</option>
                <option value="no">No</option>
            </param>
            <when value="yes">
                <conditional name="set_signature_input">
                    <param name="input_signature_choices" type="select" label="type of signatures" display="radio">
                        <option value="cosmic" selected="true">COSMIC signatures</option>
                        <option value="user_signatures">Use your own signature matrix</option>
                    </param>
                    <when value="cosmic">
                        <param name="cosmic_version" type="select" label="Version of the Cosmic signature set">
                            <option value="v2" selected="true">Cosmic v2, March 2015</option>
                            <option value="v3">Cosmic v3, May 2019</option>
                            <option value="v3.1">Cosmic v3.1, June 2020</option>
                            <option value="v3.2">Cosmic v3.2, March 2021</option>
                        </param>
                    </when>
                    <when value="user_signatures">
                        <param name="own_matrix" type="data" format="tabular"
                               label="A tab-separated matrix describing elementary signatures"
                               help="see https://cancer.sanger.ac.uk/signatures/documents/453/COSMIC_v3.2_SBS_GRCh38.txt for the required format" />
                    </when>
                </conditional>
                <param name="signum" type="integer" value="3" min="2" max="30"
                       label="selects the N most significant signatures in samples to express mutational patterns"
                       help="an integer between 2 and the number of elementary signatures in your signature matrix"/>
                <param name="display_signatures" type="select" display="radio" label="Output profiles of input signatures ?">
                    <option value="no">No</option>
                    <option value="yes" selected="true">Yes</option>
                </param>
                <param name="contrib_matrix_out" type="select" display="radio" label="Output Signature Contribution table ?"
                       help="Output the normalized signatures contributions for further visualization" >
                    <option value="no" selected="true">No</option>
                    <option value="yes">Yes</option>
                </param>
                <param name="colors" type="select" label="Color set" display="radio">
                    <option value="brewer">RColorBrewer</option>
                    <option value="intense" selected="true">Intense</option>
                </param>
            </when>
            <when value="no" />
        </conditional>
        <param name="rdata_out" type="boolean" checked="false" label="Output RData file?" help="Output all the data used by R to construct the tables and plots, can be loaded into R" />
    </inputs>
    <outputs>
        <data name="spectrum" format="pdf" label="Mutational Spectrum">
            <filter>set_spectrum['choices'] == "yes"</filter>
        </data>
        <data name="denovo" format="pdf" label="De novo signatures">
            <filter>set_denovo['choices'] == "yes"</filter>
        </data>
        <data name="sigmatrix" format="tabular" label="De novo signatures probability matrix">
            <filter>set_denovo['choices'] == "yes"</filter>
        </data>
        <data name="sig_contrib" format="pdf" label="Signature contributions">
            <filter>set_preset['choices'] == "yes"</filter>
        </data>
        <data name="sig_contrib_matrix" format="tabular" label="${tool.name}: Signature contribution table">
            <filter>set_preset['choices'] == "yes" and set_preset['contrib_matrix_out'] == "yes"</filter>
        </data>

        <data name="rdata" format="rdata" label="${tool.name}: RData file">
            <filter>rdata_out</filter>
        </data>

    </outputs>
    <tests>
         <!-- de novo signatures -->
        <test>
            <param name="vcfs">
                <collection type="list">
                    <element name="6" value="F.vcf"/>
                    <element name="7" value="G.vcf"/>
                    <element name="8" value="H.vcf"/>
                    <element name="9" value="I.vcf"/>
                </collection>
            </param>
            <param name="genome" value="BSgenome.Hsapiens.UCSC.hg38"/>
            <conditional name="set_spectrum">
                <param name="choices" value="no"/>
            </conditional>
            <conditional name="set_denovo">
                <param name="choices" value="yes"/>
            </conditional>
            <conditional name="set_preset">
                <param name="choices" value="no"/>
            </conditional>
            <param name="nrun" value="30" />
            <param name="rank" value="4" />
            <param name="newsignum" value="2" />
            <param name="cosmic_id_threshold" value="0.84" />
            <param name="rdata_out" value="true" />
            <output name="denovo" file="denovo_output1.pdf" compare="sim_size"/>
            <output name="sigmatrix" file="sigmatrix.tab" compare="sim_size"/>
            <output name="rdata" file="denovo_1.RData" compare="sim_size" delta="400000"/> <!--  delta="170000" -->
        </test>
         <!-- user defined (v3.2 restricted 30 minus 27 renamed to test nomenclature effect) -->
        <test>
            <param name="vcfs">
                <collection type="list">
                    <element name="6" value="F.vcf"/>
                    <element name="7" value="G.vcf"/>
                    <element name="8" value="H.vcf"/>
                    <element name="9" value="I.vcf"/>
                </collection>
            </param>
            <param name="genome" value="BSgenome.Hsapiens.UCSC.hg38"/>
            <param name="levels" value="FGHI_levels.tab" ftype="tabular"/>
            <conditional name="set_spectrum">
                <param name="choices" value="no"/>
            </conditional>
            <conditional name="set_denovo">
                <param name="choices" value="no"/>
            </conditional>
            <conditional name="set_preset">
                <param name="choices" value="yes"/>
                <conditional name="set_signature_input">
                    <param name="input_signature_choices" value="user_signatures" />
                    <param name="own_matrix" value="user_defined_signature.tsv" ftype="tabular"/>
                </conditional>
                <param name="contrib_matrix_out"  value="yes" />
            </conditional>
            <param name="signum" value="4" />
            <output name="sig_contrib" file="user_output.pdf" compare="sim_size"/>
            <output name="sig_contrib_matrix" file="sig_contrib_table_user_defined.tsv" compare="sim_size"/>
        </test>
         <!-- cosmic v3.2 -->
        <test>
            <param name="vcfs">
                <collection type="list">
                    <element name="6" value="F.vcf"/>
                    <element name="7" value="G.vcf"/>
                    <element name="8" value="H.vcf"/>
                    <element name="9" value="I.vcf"/>
                </collection>
            </param>
            <param name="genome" value="BSgenome.Hsapiens.UCSC.hg38"/>
            <param name="levels" value="FGHI_levels.tab" ftype="tabular"/>
            <conditional name="set_spectrum">
                <param name="choices" value="no"/>
            </conditional>
            <conditional name="set_denovo">
                <param name="choices" value="no"/>
            </conditional>
            <conditional name="set_preset">
                <param name="choices" value="yes"/>
                <conditional name="set_signature_input">
                    <param name="input_signature_choices" value="cosmic" />
                    <param name="cosmic_version" value="v3.2"/>
                </conditional>
                <param name="display_signatures"  value="no" />
                <param name="contrib_matrix_out"  value="yes" />
            </conditional>
            <param name="signum" value="3" />
            <output name="sig_contrib" file="cosmic_output_v3.pdf" compare="sim_size"/>
            <output name="sig_contrib_matrix" file="sig_contrib_table_v3.tsv" compare="sim_size"/>
        </test>
         <!-- cosmic v2 -->
        <test>
            <param name="vcfs">
                <collection type="list">
                    <element name="6" value="F.vcf"/>
                    <element name="7" value="G.vcf"/>
                    <element name="8" value="H.vcf"/>
                    <element name="9" value="I.vcf"/>
                </collection>
            </param>
            <param name="genome" value="BSgenome.Hsapiens.UCSC.hg38"/>
            <param name="levels" value="FGHI_levels.tab" ftype="tabular"/>
            <conditional name="set_spectrum">
                <param name="choices" value="no"/>
            </conditional>
            <conditional name="set_denovo">
                <param name="choices" value="no"/>
            </conditional>
            <conditional name="set_preset">
                <param name="choices" value="yes"/>
                <conditional name="set_signature_input">
                    <param name="input_signature_choices" value="cosmic" />
                    <param name="cosmic_version" value="v2"/>
                </conditional>
                <param name="contrib_matrix_out"  value="yes" />
            </conditional>
            <param name="signum" value="3" />
            <output name="sig_contrib" file="cosmic_output1.pdf" compare="sim_size"/>
            <output name="sig_contrib_matrix" file="sig_contrib_table.tsv" compare="sim_size"/>
        </test>
        <!-- simple spectrum -->
        <test>
            <param name="vcfs">
                <collection type="list">
                    <element name="1" ftype="vcf" value="G.vcf"/>
                    <element name="2" ftype="vcf" value="H.vcf"/>
                </collection>
            </param>
            <param name="genome" value="BSgenome.Hsapiens.UCSC.hg38"/>
            <conditional name="set_levels">
                <param name="choices" value="yes"/>
            </conditional>
            <param name="levels" value="GH_levels.tab" ftype="tabular"/>
            <conditional name="set_spectrum">
                <param name="choices" value="yes"/>
            </conditional>
            <conditional name="set_denovo">
                <param name="choices" value="no"/>
            </conditional>
            <conditional name="set_preset">
                <param name="choices" value="no"/>
            </conditional>
            <output name="spectrum" file="spectrum_output1.pdf" compare="sim_size"/>
        </test>
    </tests>
    <help>

**What it does**

This tool implement the R package MutationalPatterns to decompose mutations found in cancers
in a linear combination of elementary mutational signatures, as first described by
Alexandrov_ et al (2013) and recently improved in an update_ published in 2020.

.. _Alexandrov: https://www.nature.com/articles/nature12477
.. _update: https://www.nature.com/articles/s41586-020-1943-3

Sets of elementary signatures found by analyzing large cohortes of patients are published
in the COSMIC databases v2_ to v3.2_. You can choose among these COSMIC signature sets to
analyse your own VCFs.

In addition, you can use your own signature sets (or filtered COSMIC signature sets)
since MutationalPatterns implements a method to extract elementary signatures from
user-provided VCFs.

**Inputs**

* a collection of VCF files with somatic mutations calls from analysis of samples.
* a tabular table describing the correspondance of sample names to levels of a factor
  (tissues, ages, sexes, etc.)

**Outputs**

This tool returns pdf files with various visualisations of :

* the spectrum of single nucleotide mutation types
* the absolute contribution of the elementary signatures in the samples
* the relative contribution of the elementary signatures in the samples
* an unrooted hierarchical clustering of the reconstructed samples with respect to the relative contribution of elementary signatures
* the Cosine similarity between true samples and samples reconstructed with elementary signatures
* pie charts displaying the relative contribution of the n most contributing elementary signatures in the reconstructed sample profiles

.. _v2: https://cancer.sanger.ac.uk/signatures/signatures_v2/
.. _v3.2: https://cancer.sanger.ac.uk/signatures/sbs/

    </help>
    <citations>
        <citation type="doi">10.18129/B9.bioc.MutationalPatterns</citation>
        <citation type="doi">10.1186/s13073-018-0539-0</citation>
        <citation type="doi">10.1038/nature12477</citation>
        <citation type="doi">10.1038/s41586-020-1943-3</citation>
    </citations>
</tool>