<tool id="raceid_inspectclusters" name="Cluster Inspection using RaceID" version="@VERSION_RACEID@.@VERSION_PACKAGE@.1" >
    <description>examines gene expression within clusters</description>
    <macros>
        <import>macros.xml</import>
        <import>macros_inspectclusters.xml</import>
    </macros>
    <expand macro="requirements" />
    <version_command><![CDATA[
Rscript '$__tool_directory__/scripts/clusterinspect.R' @GET_VERSION@
]]></version_command>

    <command detect_errors="exit_code"><![CDATA[
#set bin = 'clusterinspect.R'
Rscript '$__tool_directory__/scripts/$bin' '$userconf' 2>&1 > '$outlog'
    ]]></command>
  
    <configfiles>
        <configfile name="userconf" ><![CDATA[
@STRING2VECTOR@

@INSPECTCLUSTERS_CHEETAH@
]]>
        </configfile>
    </configfiles>
    <inputs>
        <expand macro="inspectclusters_inputs" />
    </inputs>

    <outputs>
        <data name="outpdf" format="pdf" label="${tool.name} on ${on_string}: PDF Report" />
        <data name="outlog" format="txt" label="${tool.name} on ${on_string}: Log" >
            <filter>use_log</filter>
        </data>
    </outputs>
    <tests>
        <expand macro="inspectclusters_tests" />
    </tests>
    <help><![CDATA[

RaceID3
=========

RaceID is a clustering algorithm for the identification of cell types from single-cell RNA-sequencing data. It was specifically designed for the detection of rare cells which correspond to outliers in conventional clustering methods.

This module inspects the clusters generated from the previous clustering step (and requires the output RData file from it as input).

The tool offers three modes of inspection which can all be activated at the same time, resulting in a single PDF report:

 * Plot All Clusters:
   * Produces a tSNE of all clusters, as well as a force-directed Fruchterman-Reingold (F-R) layout which may have tidier plots.
 * Perform Subset Analysis:
   * tSNE and F-R plots with cells whose name match the specified regex highlighted
   
 * Examine Genes of Interest:
   * Expression plots highlighting a gene or genes of interest across all clusters

 * Differential Gene Testing: 
   * Examining the expression between 

 * A list of the most differentially expressed genes in each cluster
 * An output PDF that provides heatmaps for:
    * The initial and final clustering (as determined using random forest)
    * Heatmaps for each of the most differentially expressed genes in each cluster

The tool requires the RData input from the previous filtering / normalisation / confounder removal step to work.



]]></help>
    <expand macro="citations" />
</tool>