<tool id="allByAllCorr" name="Metabolite - Gene Correlation" version="@WRAPPER_VERSION@">
  <description></description>
  <macros>
    <import>macros.xml</import>
  </macros>
  <expand macro="requirements" />
  <command detect_errors="exit_code"><![CDATA[
    all_by_all_correlation.py
      -g=$geneDataset
      -gid=$geneId
      #if $geneAnnotation.useGeneAnnot == "Yes":
        -ga=$geneAnnotation.geneAnnot
        -gn=$geneAnnotation.geneAnnotName
      #end if
      -m=$metDataset
      -mid=$metId
      #if $metAnnotation.useMetAnnot == "Yes":
        -ma=$metAnnotation.metAnnot
        -mn=$metAnnotation.metAnnotName
      #end if
      -me=$method
      -t=$threshold
      -o=$output
      -c=$corMat
      -f=$figure
  ]]></command>
  <inputs>
    <param name="geneDataset" type="data" format="tabular" label="Gene Expression Wide Dataset" help="Select the Gene Expression Wide Dataset from your history"/>
    <param name="geneId" type="text" size="30" value="" label="Unique Gene FeatureID" help="Name of the Column in your Gene Expression Wide Dataset that contains unique identifiers."/>
    <conditional name="geneAnnotation">
      <param name="useGeneAnnot" type="select" label="Use Annotation File?" help="You can chose to input a file containing gene annotation information (e.g. gene names, identifiers, etc.) for labeling output files.">
        <option value="No">No</option>
        <option value="Yes">Yes</option>
      </param>
      <when value="Yes">
        <param name="geneAnnot" type="data" format="tabular" label="Gene Expression Annotation File" help="Select the Gene Expression Annotation File from your history"/>
        <param name="geneAnnotName" type="text" value="" label="Gene Labels" help="Name of the column in the Gene Expression Annotation File to use for labeling output files."/>
      </when>
    </conditional>
    <param name="metDataset" type="data" format="tabular" label="Metabolite Wide Dataset" help="Select the Metabolite Wide Dataset from your history"/>
    <param name="metId" type="text" size="30" value="" label="Unique Metabolite FeatureID" help="Name of the column in your Metabolite Wide Dataset that contains unique identifiers."/>
    <conditional name="metAnnotation">
      <param name="useMetAnnot" type="select" label="Use Annotation File?" help="You can chose to input a file containing metabolite annotation information (e.g. metabolite names, identifiers, etc.) for labeling output files.">
        <option value="No">No</option>
        <option value="Yes">Yes</option>
      </param>
      <when value="Yes">
        <param name="metAnnot" type="data" format="tabular" label="Metabolite Annotation File" help="Select the Metabolite Annotation File from your history"/>
        <param name="metAnnotName" type="text" value="" label="Metabolite Labels" help="Name of the column in the Metabolite Annotation File to use for labeling output files"/>
      </when>
    </conditional>
    <param name="method" type="select" label="Correlation Method" help="Select a correlation method.">
      <option value="pearson">Pearson</option>
      <option value="spearman">Spearman</option>
      <option value="kendall">Kendall</option>
    </param>
    <param name="threshold" type="text" value="0.05" label="P-Value threshold." help="Default: 0.05"/>
  </inputs>
  <outputs>
    <data format="tabular" name="output" label="${tool.name} on ${on_string}: Correlation File"/>
    <data format="tabular" name="corMat" label="${tool.name} on ${on_string}: Correlation Matrix"/>
    <data format="pdf" name="figure" label="${tool.name} on ${on_string}: Correlation Figure"/>
  </outputs>
  <tests>
    <test>
      <param name="geneDataset" value="gene_wide_dataset_01fhl.tsv"/>
      <param name="geneId" value="UniqueID"/>
      <param name="geneAnnot" value="gene_annotation_file_01fhl.tsv"/>
      <param name="geneName" value="GeneName"/>
      <param name="metDataset" value="metabolite_wide_dataset_01fhl.tsv"/>
      <param name="metId" value="UniqueID"/>
      <param name="metAnnot" value="metabolite_annotation_file_01fhl.tsv"/>
      <param name="metName" value="MetName"/>
      <output name="output" file="correlation_file_01fhl.tsv"/>
      <output name="corMat" file="correlation_matrix_01fhl.tsv"/>
      <output name="figure" file="correlation_figure_01fhl.pdf"/>
    </test>
  </tests>
  <help><![CDATA[

**Tool Description**

  The tool performs a correlation analysis between genes (in a gene expression wide dataset) and metabolites (in a metabolite wide dataset) 
  to generate a table of correlation coefficients. P-values for the correlation coefficients are calculated by simulating individual gene and 
  metabolite datasets 1000 times using a normal distribution with means and standard deviations generated from the data. Sample size reflects 
  the input datasets. Correlations are calculated on the simulated data. Correlations must be higher/lower than 95% of the randomly simulated 
  values to be considered significant. 

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

**Input**

  (1) **Gene Expression Dataset**  A wide formatted dataset containing measurements for each sample (where samples are in columns):

    +-----------+---------+---------+---------+-----+
    | UniqueID  | sample1 | sample2 | sample3 | ... |
    +===========+=========+=========+=========+=====+
    | gene_1    | 1.2     | 3.5     | 2.9     | ... |
    +-----------+---------+---------+---------+-----+
    | gene_2    | 1.6     | 3.2     | 3.2     | ... |
    +-----------+---------+---------+---------+-----+
    | gene_3    | 1.4     | 3.0     | 3.1     | ... |
    +-----------+---------+---------+---------+-----+
    | gene_4    | 1.6     | 2.9     | 3.1     | ... |
    +-----------+---------+---------+---------+-----+
    | ...       | ...     | ...     | ...     | ... |
    +-----------+---------+---------+---------+-----+

  (2) **Unique Gene FeatureID**  Name of the column in your Gene Expression Wide Dataset that contains unique gene identifiers.  **NOTE:** This identifier must be the gene symbol.

  (3) **Metabolomic Wide Dataset**  A wide formatted metabolomic dataset that contains measurements for each sample (where samples are in columns):

    +------------+---------+---------+---------+-----+
    | UniqueID   | sample1 | sample2 | sample3 | ... |
    +============+=========+=========+=========+=====+
    | met_1      | 10      | 20      | 10      | ... |
    +------------+---------+---------+---------+-----+
    | met_2      | 5       | 22      | 30      | ... |
    +------------+---------+---------+---------+-----+
    | met_3      | 30      | 27      | 2       | ... |
    +------------+---------+---------+---------+-----+
    | met_4      | 32      | 17      | 8       | ... |
    +------------+---------+---------+---------+-----+
    | ...        | ...     | ...     | ...     | ... |
    +------------+---------+---------+---------+-----+

  (4) **Unique Metabolite FeatureID**  Name of the column in your Metabolite Wide Dataset that contains unique metabolite identifiers.

  (5) **Annotation Files**  The user can provide (optional) Annotation Files for the Gene Expression and/or Metabolite Datasets to label the results for easier readability. The user must provide the name of the column with the desired feature name (e.g. Gene Symbol).

    +-----------+------------+-------------+-----+
    | UniqueID  | ENSEMBL_ID | Gene_Symbol | ... |
    +===========+============+=============+=====+
    | gene_1    | ENS...     | one         | ... |
    +-----------+------------+-------------+-----+
    | gene_2    | ENS...     | two         | ... |
    +-----------+------------+-------------+-----+
    | gene_3    | ENS...     | three       | ... |
    +-----------+------------+-------------+-----+
    | gene_4    | ENS...     | four        | ... |
    +-----------+------------+-------------+-----+
    | ...       | ...        | ...         | ... |
    +-----------+------------+-------------+-----+

  (6) **Correlation method**  Select the correlation coefficient to be computed from the list. Pearson, kendall, or spearman are available.

  (7) **P-Value threshold**  User specified value that limits the data in the resulting 'Correlation File' to only those correlations with P-values less than this value.

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

**Output**

  The user will obtain three outputs from the Gene - Metabolite Correlation Tool:

  (1) **Correlation File.** A file sorted by the absolute values of the correlation coeficient and including the P-value. The file contains only the correlation coefficients where the associated P-values is less than a user-specified value (default = 0.05).

    +--------+------------+-------------+-----------+
    | Gene   | Metabolite | Correlation | (p-value) |
    +========+============+=============+===========+
    | gene_1 | met_1      | 0.99        | 0.000     |
    +--------+------------+-------------+-----------+
    | gene_2 | met_4      | -0.98       | 0.000     |
    +--------+------------+-------------+-----------+
    | gene_3 | met_5      | 0.96        | 0.001     |
    +--------+------------+-------------+-----------+
    | gene_4 | met_1      | 0.95        | 0.002     |
    +--------+------------+-------------+-----------+
    | ...    | ...        | ...         | ...       |
    +--------+------------+-------------+-----------+

  (2) **Correlation Matrix.** Output correlation matrix.

    +--------+-------+-------+-------+-------+-----+
    | Gene   | met_1 | met_2 | met_3 | met_4 | ... |
    +========+=======+=======+=======+=======+=====+
    | gene_1 | 0.99  | 0.56  | 0.25  | 0.33  | ... |
    +--------+-------+-------+-------+-------+-----+
    | gene_2 | -0.57 | 0.63  | -0.14 | 0.01  | ... |
    +--------+-------+-------+-------+-------+-----+
    | gene_3 | 0.62  | 0.96  | 0.20  | 0.32  | ... |
    +--------+-------+-------+-------+-------+-----+
    | gene_4 | 0.95  | 0.25  | 0.16  | 0.44  | ... |
    +--------+-------+-------+-------+-------+-----+
    | ...    | ...   | ...   | ...   | ...   | ... |
    +--------+-------+-------+-------+-------+-----+

  (3) **Correlation Figure.** Network representation of the top 500 gene-metabolite correlations based on the absolute value of the correlation coeficients. Maximum number of correlations in the netork is 500.

  ]]>
  </help>
  <citations>
    <citation type="bibtex">@article{dejean2013mixomics,
    title={mixOmics: Omics data integration project},
    author={Dejean, Sebastien and Gonzalez, Ignacio and L{\^e} Cao, Kim-Anh and Monget, Pierre and Coquery, J and Yao, F and Liquet, B and Rohart, F},
    journal={R package},
    year={2013}
    }</citation>
    <citation type="bibtex">@ARTICLE{Kirpich17secimtools,
    author = {Alexander S. Kirpich, Miguel Ibarra, Oleksandr Moskalenko, Justin M. Fear, Joseph Gerken, Xinlei Mi, Ali Ashrafi, Alison M. Morse, Lauren M. McIntyre},
    title = {SECIMTools: A suite of Metabolomics Data Analysis Tools},
    journal = {BMC Bioinformatics},
    year = {in press}
    }</citation>
    <citation type="bibtex">
    @article{garcia2010paintomics,
    title={Paintomics: a web based tool for the joint visualization of transcriptomics and metabolomics data},
    author={Garc{\'\i}a-Alcalde, Fernando and Garc{\'\i}a-L{\'o}pez, Federico and Dopazo, Joaqu{\'\i}n and Conesa, Ana},
    journal={Bioinformatics},
    volume={27},
    number={1},
    pages={137--139},
    year={2010},
    publisher={Oxford University Press}
    }</citation>
  </citations>
</tool>