comparison homogeneity_normality.xml @ 0:9f679060051a draft

"planemo upload for repository https://github.com/Marie59/Data_explo_tools commit 2f883743403105d9cac6d267496d985100da3958"

0:9f679060051a

<tool id="ecology_homogeneity_normality" name="Homoscedasticity and normality" version="@VERSION@" profile="20.01">
    <description>Checks the homogeneity of the variance and the normality of the distribution
    </description>
    <macros>
        <import>macro.xml</import>
    </macros>
    <expand macro="Explo_requirements">
        <requirement type="package" version="3.0_10">r-car</requirement>
        <requirement type="package" version="1.1.1">r-cowplot</requirement>
        <requirement type="package" version="0.4.0">r-ggpubr</requirement>
    </expand>
    <command detect_errors="exit_code"><![CDATA[
        Rscript
            '$__tool_directory__/graph_homogeneity_normality.r'
            '$input'
            '$colnames'
            '$date'
            '$species'
            '$variable'
            '$output_levene'
            '$output_ks'
            '$plots'
        ]]>
    </command>
    <inputs>
        <expand macro="explo_input"/>
        <param name="date" label="Select column containing temporal data (year, date, ...)" type="data_column" data_ref="input" use_header_names="true"/>
        <param name="species" label="Select column containing species" type="data_column" data_ref="input" use_header_names="true"/>
        <param name="variable" label="Select column containing numerical values (like abundances)" type="data_column" data_ref="input" numerical="true" use_header_names="true"/>
    </inputs>
    <outputs>
        <data name="output_levene" from_work_dir="levene.txt" format="txt" label="Levene test - Homoscedasticity"/>
        <data name="output_ks" from_work_dir="ks.txt" format="txt" label="Kolmogorov-Smirnov test - Normality"/>
        <collection type="list" name="plots">
            <discover_datasets pattern="(?P&lt;designation&gt;.+)\.png" visible="false" format="png"/>
        </collection>
    </outputs>
    <tests>
        <test>
            <param name="input" value="data_frenchbbs.tabular"/>
            <param name="colnames" value="true"/>
            <param name="date" value="2"/>
            <param name="species" value="3"/>
            <param name="variable" value="4"/>
            <output name="output_levene" value="levene.txt"/>
            <output name="output_ks">
                <assert_contents>
                    <has_n_lines n="9"/>
                </assert_contents>
            </output>
            <output_collection name="plots" type="list" count="2">
                <element name="Homogeneity" ftype="png">
                    <assert_contents>
                        <has_text text="PNG"/>
                    </assert_contents>
                </element>
                <element name="Normal_distribution" ftype="png">
                    <assert_contents>
                        <has_text text="PNG"/>
                    </assert_contents>
                </element>
            </output_collection>
        </test>
    </tests>
    <expand macro="topic"/>
    <help><![CDATA[
===============================================================
Checks homogeneity of variance (homoscedasticity) and normality
===============================================================


**What it does**


This tool checks homogeneity of variances (Levene test) for every species and represents it through multiple boxplots and the normal distribution (Kolmogorov-Smirnov test) represented by a distribution histogram and a Q-Q plot.

If the levene test is significant (P-value in column Pr < 0.5 and at least one * at the end of the 4th line), variances aren't homogeneous, the hypothesis of *homoscedasticity is rejected*.

If the K-S test is significant (p-value < 0.5), your numerical variable *isn't* normally distributed, the hypothesis of *normality is rejected*.


**Input description**

A tabular file with observation data. Must at least contain three columns temporal referential (year, date, ...), species and a numerical factor.

+----------+-----------+-----------+
|   date   |   number  |  species  |
+==========+===========+===========+
|   year   |     4     | speciesID |
+----------+-----------+-----------+
|  ...     |    ...    |    ...    |
+----------+-----------+-----------+

**Output**

A text output that summarizes the result of the performed Levene test for homogeneity of variances (homoscedasticity).

A text output that summarizes the result of the performed Kolmogrov-Smirnov test for normality.

Multiple PNG files representing the homogeneity of variances for each species at each time point of the study.

One PNG file representing the distribution of the data.
    ]]>    </help>
    <expand macro="explo_bibref"/>

</tool>