<tool id="quantp" name="QuanTP" version="1.0.0">
    <description>Correlation between protein and transcript abundances</description>
    <requirements>
        <requirement type="package" version="1.10.4">r-data.table</requirement>
        <requirement type="package" version="3.0.1">r-gplots</requirement>
        <requirement type="package" version="0.7.6">r-dplyr</requirement>
        <requirement type="package" version="3.0.0">r-ggplot2</requirement>
        <requirement type="package" version="0.4.5">r-ggfortify</requirement>
    </requirements>
    <command detect_errors="exit_code"><![CDATA[
Rscript '$__tool_directory__/quantp.r' 
    $experiment_design_option.sample_type 
    $experiment_design_option.method_type 
    $experiment_design_option.exp_design 
    '$pe_exp' 
    '$te_exp' 
    $experiment_design_option.correction_method 
    $cooksd_upper 
    $nclust 
    $hm_nclust 
    $experiment_design_option.volcano_with 
    '$html_file' 
    '$html_file.files_path'
]]></command>
    <inputs>
        <param name="pe_exp" type="data" format="tabular" label="Input Protein Abundance File" help="Protein abundance input file"/>
        <param name="te_exp" type="data" format="tabular" label="Input RNA Abundance File" help="Transcript abundance input file"/>
        <conditional name="experiment_design_option">
            <param name="sample_type" type="select" label="Select data input type" help="If the input files already have fold-change values, select Log fold-change. Else, select abundances and the tool will perform the fold-change analysis">
                <option value="multiple" selected="True">Abundances from different conditions with or without replicates (in multiple columns)</option>
                <option value="logfold">Log fold-change values (or single condition abundance without replicates in single column) data</option>
            </param>
            <when value="logfold">
                <param name="exp_design" type="hidden" value="none" />
                <param name="method_type" type="hidden" value="none" />
                <param name="correction_method" type="hidden" value="none" />
                <param name="volcano_with" type="hidden" value="pval" />
            </when>
            <when value="multiple">
                <param name="exp_design" type="data" format="tabular" help="Please check the format of the experiment design file">
                    <label>Experiment design File (Please see the format below)</label>
                </param>
                <param name="method_type" type="select" label="Data summarization method" help="Perform T-Test on selecting Mean; Wilcoxon Ranksum Test on selecting Median">
                    <option value="mean" selected="True">Mean (Default)</option>
                    <option value="median">Median</option>
                </param>
                <param name="correction_method" type="select" label="Multiple testing correction method">
                    <option value="BH" selected="True">Benjamini and Hochberg (BH) (Default)</option>
                    <option value="holm">Holm</option>
                    <option value="hochberg">Hochberg</option>
                    <option value="hommel">Hommel</option>
                    <option value="bonferroni">Bonferroni</option>
                    <option value="BY">Benjamini and Yekutieli (BY)</option>
                    <option value="none">None</option>
                </param>
                <param name="volcano_with" type="select" display="radio" label="Volcano plot with p-value or adjusted p-value">
                    <option value="pval" selected="True">P-value (Default)</option>
                    <option value="adj_pval">Adjusted P-value</option>
                </param>
            </when>
        </conditional>
        <param name="cooksd_upper" type="integer" value="4" optional="false" >
            <label>Influential Observation cutoff: Observations > value * mean of Cook's distances (Default: "4" * mean(Cook's Distance))</label>
        </param>
        <param name="nclust" type="select" label="K-mean clustering: Number of clusters">
            <option value="1">1</option>
            <option value="2">2</option>
            <option value="3">3</option>
            <option value="4" selected="True">4</option>
            <option value="5">5</option>
            <option value="6">6</option>
            <option value="7">7</option>
            <option value="8">8</option>
            <option value="9">9</option>
            <option value="10">10</option>
        </param>
      
        <param name="hm_nclust" type="select" label="Hierarchical clustering: Number of clusters (from Heatmap)">
            <option value="1">1</option>
            <option value="2">2</option>
            <option value="3">3</option>
            <option value="4">4</option>
            <option value="5" selected="True">5</option>
            <option value="6">6</option>
            <option value="7">7</option>
            <option value="8">8</option>
            <option value="9">9</option>
            <option value="10">10</option>
        </param>
    </inputs>
    <outputs>
        <data format="html" name="html_file" label="protein transcript correlation on ${pe_exp.name} and ${te_exp.name}"/>
    </outputs>
    <tests>
        <test>
            <conditional name="experiment_design_option">
                <param name="sample_type" value="multiple"/>
                <param name="method_type" value="mean"/>
                <param name="exp_design" value="exp_design_file.tabular" ftype="tabular" />
                <param name="correction_method" value="BH"/>
                <param name="volcano_with" value="pval"/>
            </conditional>
            <param name="pe_exp" value="protein_data.tabular" ftype="tabular" />
            <param name="te_exp" value="transcript_data.tabular" ftype="tabular" />
            <param name="cooksd_upper" value="4"/>
            <param name="nclust" value="4"/>
            <param name="hm_nclust" value="5"/>
            <output name="html_file">
                <assert_contents>
                    <has_text text="SAMPLE DISTRIBUTION" />
                </assert_contents>
            </output>
        </test>
    </tests>
    <help><![CDATA[

**What it does**

QuanTP correlates *transcript abundance* and *protein abundance* to examine the association between them.

It either takes in the log fold-change of abundances as input or raw abundances from different conditions where it calculates the log ratios of abundances between two conditions.

Transcript input file can be generated from the quantitative RNA-Seq study whereas Protein input file can be generated from quantitative analysis of mass-spectrometry-based protein data.

-----

**Input file formats**

**Protein data file**

First column - Gene

Following columns - Abundance values (or log fold-change values)

Example of Protein input file

====== ========= ========= ========= ========= ========= ========= ========= =========
Gene   sample1   sample2   sample3   sample4   sample5   sample6   sample7   sample8
------ --------- --------- --------- --------- --------- --------- --------- ---------
GeneX  value     value     value     value     value     value     value     value    
GeneY  value     value     value     value     value     value     value     value    
GeneZ  value     value     value     value     value     value     value     value    
====== ========= ========= ========= ========= ========= ========= ========= =========


**Transcript data file**

First column - Gene

Following columns - Abundance values (or log fold-change values)

Example of Transcript input file

====== ========= ========= ========= ========= ========= ========= ========= =========
Gene   sample1   sample2   sample3   sample4   sample5   sample6   sample7   sample8 
------ --------- --------- --------- --------- --------- --------- --------- ---------
GeneX  value     value     value     value     value     value     value     value    
GeneY  value     value     value     value     value     value     value     value    
GeneZ  value     value     value     value     value     value     value     value    
====== ========= ========= ========= ========= ========= ========= ========= =========


**Data input type**

If data input type is abundance, experiment design file is required.

Example of experiment design file

======== =========
case     groupA   
control  groupB   
sample1  groupA   
sample2  groupA   
sample3  groupA   
sample4  groupA   
sample5  groupB   
sample6  groupB   
sample7  groupB   
sample8  groupB   
======== =========

Note: No title/header in experiment design file and the first two lines of the experiment design must have keyword "case" and "control"

  ]]>
  </help>
  <citations>
        <citation type="bibtex">
@misc{QuanTP: A software resource for quantitative proteo-transcriptomic comparative data analysis and informatics,
    author={Praveen Kumar, Priyabrata Panigrahi, James Johnson, Wanda Weber, Subina Mehta, Ray Sajulga, Caleb Easterly, Brian Crooker, Mohammad Heydarian, Krishanpal Anamika, Timothy Griffin, and Pratik Jagtap},
    year={2018},
    title={QuanTP}
}
        </citation>
    </citations>
</tool>