annotate xml_intensity_check.xml @ 0:860907bf0e0f draft default tip

planemo upload for repository https://github.com/your_org_or_user/tools-metabolomics commit 3dd90b60510fa72eb8351464ec5d5883b0b6222a

<tool id="intensity_check" name="Intensity Check" version="@TOOL_VERSION@" profile="22.05">
    <description>Statistical measures, number of missing values and mean fold change</description>
    <macros>
        <token name="@TOOL_VERSION@">2.0.0</token>
    </macros>

    <edam_topics>
        <edam_topic>topic_3172</edam_topic>
    </edam_topics>
    <edam_operations>
        <edam_operation>operation_2428</edam_operation>
        <edam_operation>operation_2428</edam_operation>
    </edam_operations>

    <requirements>
        <requirement type="package" version="4">r-base</requirement>
        <requirement type="package" version="2.2.5">r-argparse</requirement>
    </requirements>

    <required_files>
        <include path="wrapper_intensity_check.R"/>
    </required_files>

    <command detect_errors="aggressive"><![CDATA[
Rscript '$__tool_directory__/wrapper_intensity_check.R'
    --dataMatrix_in '$dataMatrix_in'
    --sampleMetadata_in '$sampleMetadata_in'
    --variableMetadata_in '$variableMetadata_in'
    --method '${method_cond.method}'
    --chosen_stat '${method_cond.chosen_stat}'
#if $method_cond.method == "each_class"
    --class_col '${method_cond.class_col}'
    --test_fold '${method_cond.test_fold_cond.test_fold}'
    #if $method_cond.test_fold_cond.test_fold == "Yes"
        --logarithm '${method_cond.test_fold_cond.logarithm}'
    #end if
#end if
#if $method_cond.method == "one_class"
    --class_col '${method_cond.class_col}'
    --class1 '${method_cond.class1}'
    --test_fold '${method_cond.test_fold_cond.test_fold}'
    #if $method_cond.test_fold_cond.test_fold == "Yes"
        --fold_frac '${method_cond.test_fold_cond.fold_frac}'
        --logarithm '${method_cond.test_fold_cond.logarithm}'
    #end if
#end if
    --variableMetadata_out '$variableMetadata_out'
    --graphs_out '$graphs_out'
    ]]></command>

    <inputs>
        <param name="dataMatrix_in" type="data" format="tabular,tsv" label="Data matrix file"
            help="Main data matrix. Rows are variables (features), columns are samples. The first column must contain variable identifiers, and the first row must contain sample identifiers." />
        <param name="sampleMetadata_in" type="data" format="tabular,tsv" label="Sample metadata file"
            help="Sample metadata table. Each row corresponds to a sample, and columns provide sample annotations (e.g., class, batch, group). The first column must contain sample identifiers matching the data matrix columns." />
        <param name="variableMetadata_in" type="data" format="tabular,tsv" label="Variable metadata file"
            help="Variable metadata table. Each row corresponds to a variable (feature), and columns provide variable annotations (e.g., m/z, retention time). The first column must contain variable identifiers matching the data matrix rows." />
        <conditional name="method_cond">
            <param name="method" type="select" label="Computation method"
                help="Choose how to group samples for the calculation of statistics and fold changes. 
                • Without distinction between samples: All samples are treated as a single group. 
                • For each class of samples: Samples are grouped by a class column in the sample metadata. 
                • Between one class and all the remaining samples: Focus on one class versus all others.">
                <option value="no_class">Without distinction between samples</option>
                <option value="each_class">For each class of samples</option>
                <option value="one_class">Between one class and all the remaining samples</option>
            </param>
            <when value="no_class">
                <param name="chosen_stat" type="select" display="checkboxes" multiple="true" label="Statistics"
                    help="Select one or more statistics to calculate for each variable across all samples. At least one must be selected.">
                    <option value="mean">Mean</option>
                    <option value="sd">Standard deviation</option>
                    <option value="median">Median</option>
                    <option value="quartile">Quartile</option>
                    <option value="decile">Decile</option>
                    <option value="NA">Missing values</option>
                    <validator type="no_options" message="Please choose at least one statistic representation" />
                </param>
            </when>
            <when value="each_class">
                <param name="class_col" type="data_column" data_ref="sampleMetadata_in" use_header_names="true"
                    label="Class column"
                    help="Select the column in the sample metadata file that defines the sample classes (groups) for statistics and fold change calculations." />
                <param name="chosen_stat" type="select" display="checkboxes" multiple="true" label="Statistics"
                    help="Select one or more statistics to calculate for each variable within each class. Leave blank if you only want to calculate mean fold change.">
                    <option value="mean">Mean</option>
                    <option value="sd">Standard deviation</option>
                    <option value="median">Median</option>
                    <option value="quartile">Quartile</option>
                    <option value="decile">Decile</option>
                    <option value="NA">Missing values</option>
                </param>
                <conditional name="test_fold_cond">
                    <param name="test_fold" type="select" display="radio" label="Calculate the mean fold change"
                        help="Select 'Yes' to calculate the mean fold change between all pairs of classes.">
                        <option value="Yes">Yes</option>
                        <option value="No" selected="true">No</option>
                    </param>
                    <when value="Yes">
                        <param name="logarithm" type="select" label="Logarithm"
                            help="Choose a logarithmic transformation for the mean fold change values. 
                            • none: No transformation 
                            • log2: Log2 transformation 
                            • log10: Log10 transformation">
                            <option value="none" selected='true'>none</option>
                            <option value="log2">log2</option>
                            <option value="log10">log10</option>
                        </param>
                    </when>
                    <when value="No">
                    </when>
                </conditional>
            </when>
            <when value="one_class">
                <param name="class_col" type="data_column" data_ref="sampleMetadata_in" use_header_names="true"
                    label="Class column"
                    help="Select the column in the sample metadata file that defines the sample classes (groups). The selected class will be compared to all other samples." />
                <param name="class1" type="text" label="Selected class"
                    help="Enter the name of the class to compare against all other samples. This class will be used as numerator or denominator in fold change calculations." />
                <param name="chosen_stat" type="select" display="checkboxes" multiple="true" label="Statistics"
                    help="Select one or more statistics to calculate for the selected class and the remaining samples. Leave blank if you only want to calculate mean fold change.">
                    <option value="mean">Mean</option>
                    <option value="sd">Standard deviation</option>
                    <option value="median">Median</option>
                    <option value="quartile">Quartile</option>
                    <option value="decile">Decile</option>
                    <option value="NA">Missing values</option>
                </param>
                <conditional name="test_fold_cond">
                    <param name="test_fold" type="select" display="radio" label="Calculate the mean fold change"
                        help="Select 'Yes' to calculate the mean fold change between the selected class and all other samples.">
                        <option value="Yes">Yes</option>
                        <option value="No">No</option>
                    </param>
                    <when value="Yes">
                        <param name="fold_frac" type="select" label="Where should the class be placed for the mean fold change calculation?" display="radio"
                            help="Choose whether the selected class should be used as the numerator (Top) or denominator (Bottom) in the fold change calculation.">
                            <option value="Top">Numerator (Top)</option>
                            <option value="Bottom">Denominator (Bottom)</option>
                        </param>
                        <param name="logarithm" type="select" label="Logarithm"
                            help="Choose a logarithmic transformation for the mean fold change values. 
                            • none: No transformation 
                            • log2: Log2 transformation 
                            • log10: Log10 transformation">
                            <option value="none">none</option>
                            <option value="log2">log2</option>
                            <option value="log10">log10</option>
                        </param>
                    </when>
                    <when value="No">
                    </when>
                </conditional>
            </when>
        </conditional>
    </inputs>

    <outputs>
        <data name="variableMetadata_out" format="tabular" label="IC_${variableMetadata_in.name}" />
        <data name="graphs_out" format="pdf" label="IC_Graphs" />
    </outputs>

    <tests>
        <test>
            <param name="dataMatrix_in" value="ic_in_dataMatrix.tabular" />
            <param name="sampleMetadata_in" value="ic_in_sampleMetadata.tabular" />
            <param name="variableMetadata_in" value="ic_in_variableMetadata.tabular" />
            <conditional name="method_cond">
                <param name="method" value="no_class" />
                <param name="chosen_stat" value="mean,sd,median,quartile,decile,NA" />
            </conditional>
            <output name="variableMetadata_out" value="ic_out_variableMetadata.tabular" />
            <output name="graphs_out" value="ic_out_graphs.pdf" />
        </test>
        <test>
            <param name="dataMatrix_in" value="ic_in_dataMatrix.tabular" />
            <param name="sampleMetadata_in" value="ic_in_sampleMetadata.tabular" />
            <param name="variableMetadata_in" value="ic_in_variableMetadata.tabular" />
            <conditional name="method_cond">
                <param name="chosen_stat" value="mean,sd,median,quartile,decile,NA" />
                <param name="method" value="each_class" />
                <param name="class_col" value="2" />
            </conditional>
            <output name="variableMetadata_out" value="ic_out_variableMetadata_classes.tabular" />
            <output name="graphs_out" value="ic_out_graphs_classes.pdf" />
        </test>
    </tests>

    <help><![CDATA[

.. class:: infomark

**Author:**  
Anthony Fernandes for original code (PFEM - INRA)  
**Maintainer:** Melanie Petera (PFEM - INRAE - MetaboHUB)

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

====================
Intensity Check
====================

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

This tool performs various metrics: mean fold change calculation, number and proportion of missing values, and mean, sd and decile calculation.
You can choose to perform these metrics according to sample groups defined in the sample metadata file given as input.

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

.. image:: int_check.png
   :width: 800

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

+----------------------------+------------+
| Parameter                  |   Format   |
+============================+============+
| 1 : Data matrix file       |   tabular  |
+----------------------------+------------+
| 2 : Sample metadata file   |   tabular  |
+----------------------------+------------+
| 3 : Variable metadata file |   tabular  |
+----------------------------+------------+

----------

Parameters
----------

**Computation method**  
    - **Without distinction between samples:** calculates chosen statistic(s) for each variable.  
    - **For each class of samples:** separates samples between each class (class column to be specified). Chosen statistic(s) and/or mean fold change are calculated for each of them.  
    - **Between one class versus all the remaining samples:** If you want to focus only on one class versus all the remaining samples without class distinction.  

    In the case of two classes: "each class" and "one class" give the same results for statistical measures. We recommend to choose "one class" for mean fold change calculation in order to select the class you want to put as numerator or denominator (see below).

**Statistics**  
    Select the statistical measures you want to add in the variable metadata table. If the method is "each class" or "one class", you can choose to leave this section blank if you only want to calculate the mean fold change (see below).

**Class column** (*only if "each class" or "one class"*)  
    Select the class column in sample metadata table.

**Selected class** (*only if "one class"*)  
    If the method is "one class", specify it. Remaining samples will be named "Other".

**Calculate the mean fold change** (*only if "each class" or "one class"*)  
    Choose if you want to calculate the mean fold change. If the method is "each class": mean fold change will be calculated for all combinations of classes. If the method is "one class": it will be calculated between the selected class (see above) and the remaining samples.

**Where should the class be placed for the mean fold change calculation?** (*only if "one class"*)  
    If the method is "one class", choose "top" or "bottom" to put the selected class as numerator or denominator (respectively) for the mean fold change calculation.

**Logarithm** (*only if "each class" or "one class"*)  
    Choose if you want to transform the mean fold change with a log2 or log10.

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

**Variable metadata file**  
    Contains the previous columns in variable metadata file and the new ones.  
    In the column names for fold, the first class specified is the one used as numerator in the ratio.

**Graphs file**  
    Contains barplots with the proportion of NA considering classes and boxplots with the fold values.

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

----------------------
Additional information
----------------------

.. class:: warningmark

For more information about input files, refer to the corresponding "W4M HowTo" page:  
`W4M table format for Galaxy <https://nextcloud.inrae.fr/s/qLkNZRf84QQ5YLY>`_

    ]]></help>
    <citations>
        <citation type="doi">10.1002/cpz1.70095</citation>
    </citations>
</tool>