<tool id="pathifier" name="Pathifier" version="@TOOL_VERSION@+galaxy@VERSION_SUFFIX@" profile="@PROFILE@">
    <description>: Quantify deregulation of pathways in cancer</description>
    <macros>
        <token name="@TOOL_VERSION@">1.40.0</token>
        <token name="@VERSION_SUFFIX@">0</token>
        <token name="@PROFILE@">23.0</token>
    </macros>
    <requirements>
        <requirement type="package" version="@TOOL_VERSION@">bioconductor-pathifier</requirement>
        <requirement type="package" version="1.7.4">r-optparse</requirement>
        <requirement type="package" version="1.0.12">r-pheatmap</requirement>
        <requirement type="package" version="0.3_44">r-scatterplot3d</requirement>
        <requirement type="package" version="0.4.15">r-circlize</requirement>
    </requirements>
    <stdio>
        <exit_code range="1:" level="fatal" description="Tool exception" />
    </stdio>
    <command detect_errors="exit_code"><![CDATA[ 
        Rscript '$__tool_directory__/pathifier.R' 
            --exp '$input'
            --sep '$input_sep'
            --genes '$genes'

            #if str($reference.reference_selector) == 'TRUE':
                --is_normal '$reference.reference_selector'
                --normals '$reference.normals'
            #end if

            --max_stability '$max_stability'
            --attempts '$attempts'
            --min_std '$min_std'
            --min_exp '$min_exp'

            --heatmap_cluster_cells '$heatmap_cluster_cells'
            --heatmap_cluster_pathways '$heatmap_cluster_pathways'
            --heatmap_show_cell_labels '$heatmap_show_cell_labels'
            --heatmap_show_pathway_labels '$heatmap_show_pathway_labels'

            --pds '$pds'
            --logfile '$logfile'
            --plot '$plot'
            --rdata '$rdatafile'
]]></command>
    <inputs>
        <param name="input" type="data" format="txt,tabular" label="expression data"/>
        <param name="input_sep" type="select" label="Input column separator">
            <option value="tab" selected="true">Tabs</option>
            <option value=",">Comma</option>
        </param>
        <param name="genes" type="data" format="txt" label="Gene sets Pathways" 
               help="Must be in gmt format (one pathway per line : Name, description, genes (one by column), tab separated)" />
        <conditional name="reference">
            <param name="reference_selector" label="Do you have non cancer transcriptomes in your data set ?" type="select"
                   help="Yes, if the starting curve depends on the matrix of points with in a certain row order
                         (first 'normal' then 'cancer' samples), otherwise (No) the first principal component is used. See help section for more informations">
                <option selected="True" value="TRUE">Yes</option>
                <option value="">No</option>
            </param>
            <when value="TRUE">
                <param name="normals" type="data" format="tabular" label="Sample status"
                       help="A two-column data frame, first column contains data labels, second column the levels of sample status : 1 = Healthy, 0 = Tumor (no header)" />
            </when>
            <when value="">
            </when>
        </conditional>
        <param name="max_stability" label="Throw away components leading to low stability of sampling noise" type="boolean" truevalue="TRUE" 
               falsevalue="FALSE" checked="true" />
        <param name="attempts" type="integer" label="Number of runs to determine stability" value="100"/>
        <param name="min_std" type="text" value="0.4" label="Minimum of standard deviation to filter out low variable genes"
               help="Use 'data' to use the minimum standard deviation of your data" />
        <param name="min_exp" type="text" value="4" label="Minimum of gene expression to filter out low variable genes"
               help="Use 'data' to use the minimum expression of your data" />
        <param name="heatmap_cluster_cells" label="Cluster samples on heatmap" type="boolean" truevalue="TRUE" falsevalue="FALSE" checked="true" />
        <param name="heatmap_cluster_pathways" label="Cluster pathways on heatmap" type="boolean" truevalue="TRUE" falsevalue="FALSE" checked="true" />
        <param name="heatmap_show_cell_labels" label="Show sample labels on heatmap" type="boolean" truevalue="TRUE" falsevalue="FALSE" checked="false" />
        <param name="heatmap_show_pathway_labels" label="Cluster pathway labels on heatmap" type="boolean" truevalue="TRUE" falsevalue="FALSE" checked="true" />
        <param label="Return log file of pathifier" name="log" type="select" >
            <option value="no" selected="True">No</option>
            <option value="yes">Yes</option>
        </param>
        <param label="Return Pathifier S4 object" name="rdata" type="select" >
            <option value="no" selected="True">No</option>
            <option value="yes">Yes</option>
        </param>
    </inputs>
    <outputs>
        <data name="pds" format="tabular" label="Pathifier Deregulation Score (PDS) of ${on_string}" />
        <data name="logfile" format="txt" label="Pathifier log file of ${on_string}" >
            <filter>log == 'yes'</filter>
        </data>
        <data name="plot" format="pdf" label="Pathifier vizualization of ${on_string}" />
        <data name="rdatafile" format="rdata" label="Pathifier S4 object of ${on_string}" >
            <filter>rdata == 'yes'</filter>
        </data>
    </outputs>
    <tests>
        <test expect_num_outputs="3">
            <param name="input" value="sheffer.tsv" ftype="tabular"/>
            <param name="genes" value="kegg_pathways.gmt" ftype="txt" />
            <param name="reference_selector" value="TRUE" />
            <param name="normals" value="normals.tsv" ftype="tabular" />
            <param name="log" value="yes" />
            <param name="attempts" value="100" />
            <output name="logfile" file="sheffer.kegg.log" ftype="txt" compare="sim_size" />
            <output name="pds" file="sheffer.kegg.tsv" ftype="tabular"/>
            <output name="plot" file="plot.pdf" ftype="pdf" compare="sim_size" />
        </test>
        <test expect_num_outputs="3">
            <param name="input" value="sheffer_noref.tsv" ftype="tabular"/>
            <param name="genes" value="kegg_pathways.gmt" ftype="txt" />
            <param name="reference_selector" value="" />
            <param name="log" value="no" />
            <param name="rdata" value="yes" />
            <param name="attempts" value="50" />
            <param name="min_exp" value="data" />
            <param name="min_std" value="data" />
            <output name="pds" file="sheffer.kegg_noref.tsv" ftype="tabular"/>
            <output name="rdatafile" file="sheffer.kegg_noref.rdata" ftype="rdata" compare="sim_size" />
            <output name="plot" file="plot_noref.pdf" ftype="pdf" compare="sim_size" />
        </test>
    </tests>
    <help>

**What it does**

Pathifier is an algorithm that infers pathway deregulation scores for each (tumor) sample on the basis
of expression data. This score is determined, in a context-specific manner, for every particular dataset
and type of cancer that is being investigated. The algorithm  transforms gene-level information into
pathway-level information, generating a compact and biologically relevant representation of each sample.

For each pathway analysed, the transcriptome datasets are plotted in the gene pathway space using a
Principal Component analysis (PCA) and a principal curve is regressed from these points. All transcriptomes 
are projected on the nearest point of this curve. Pathifier finally computes a score which corresponds to 
the distance (normalized to 1) of each point (transcriptome) to the curve origin. 

**Inputs**

    * a matrix of n columns of observations (generally RNAseq experiments) and k rows of variables (generally k genes).
    * a Gene Matrix Transposed file (GMT format) where each row represents a gene set :
        * first column : gene set name (pathway name)
        * second : description of gene set
        * third and + : list of genes that composed the gene set tab-separated  
    * (Optional) a two column table with no header, to described transcriptome status (Tumor or not) : 
        * first column : sample labels
        * second : levels of sample status : 1 = Healthy, 0 = Tumor

**Outputs**

    * Table of Pathway Deregulation Scores : one by pathway (column) and by transcriptome (row)
    * Visualization of PDS (pdf) : 
        * Principal curve of different PDS 
        * Heatmap of PDS that allows (through clustering) to see pattern in pathway deregulation
    * (Optional) Log file of Pathifier algorithm
    * (Optional) Pathifier S4 object which contains all informations and results generated by Pathifier, for each pathway :
        * `scores` : PDS scores
        * `genesinpathway` : Gene identifiers in each pathway
        * `newmeanstd`
        * `origmeanstd`
        * `pathwaysize` : Number of genes retained in pathway
        * `curves` : Coordinates of transcriptomes projected on the principal curve
        * `curves_order` : Order of transcriptomes along the principal curve
        * `z` : z-scores matrix
        * `compin`
        * `xm`
        * `xs`
        * `center`
        * `rot`
        * `pctaken` : Number of principal component retained for pathifier analysis
        * `samplings`
        * `sucess` : List of pathway index that passed Pathfiier filters (more info in log file)
        * `logfile` : Name of logfile

    </help>
    <citations>
        <citation type="bibtex">@Manual{,
            title = {{pathifier}: Quantify deregulation of pathways in cancer},
            author = {Yotam Drier},
            year = {2013-06-27},
            note = {R package version 1.22.0},
            url = {https://git.bioconductor.org/packages/pathifier},
            }
        </citation>
        <citation type="bibtex">@article {Drier6388,
            author = {Drier, Yotam and Sheffer, Michal and Domany, Eytan},
            title = {Pathway-based personalized analysis of cancer},
            volume = {110},
            number = {16},
            pages = {6388--6393},
            year = {2013},
            doi = {10.1073/pnas.1219651110},
            publisher = {National Academy of Sciences},
            issn = {0027-8424},
            URL = {https://www.pnas.org/content/110/16/6388},
            eprint = {https://www.pnas.org/content/110/16/6388.full.pdf},
            journal = {Proceedings of the National Academy of Sciences}
            }
        </citation>
  </citations>
</tool>