<tool id="segmentation" name="Segmentation and calling" version="1.2.0">
  <description>of a previously normalized signal</description>
   <requirements>
        <container type="docker">sblanck/mpagenomicsdependencies</container>
  </requirements>
  <command>
    <![CDATA[ 
        Rscript 
        ${__tool_directory__}/segmentation.R 
  		#if $signalType.signal == "CN":
  			--nbcall '$signalType.nbcall' 
  			--cellularity '$signalType.cellularity'
  		#else
  			--nbcall '3' 
  			--cellularity '1.0'
  		#end if
  		--input '$input' 
  		--new_file_path '$output.extra_files_path' 
  		--outputlog '$outputlog' 
  		--output '$output' 
  		--log '$log' 
  		--outputgraph '$outputgraph'
  		--graph '$graph' 
  		--method '$method' 
  		--signalType '$signalType.signal'
  		--user_id '$__user_id__'
  	]]>
  		
  </command>
  <inputs>
    <param name="input" type="data" format="sef" label="Input Signal" help="see below for more information on file format"/>
	    
    <param name="method" type="select" label="Segmentation method" help="">
      <option value="cghseg">cghseg</option>
      <option value="PELT">PELT</option>
    </param>
    
	<conditional name="signalType">
    	<param name="signal" type="select" multiple="false" label="Signal type">
     		<option value="CN">CN</option>
      		<option value="fracB">fracB</option>
    	</param> 
    	<when value="fracB"/>
    	<when value="CN">
    		
    <param name="nbcall" type="select" label="Number of calling classes" help="The number of levels to be used for calling. Either 3 (loss, normal, gain), 4 (including amplifications), 5 (including double deletions) ">
      <option value="3">3</option>
      <option value="4">4</option>
      <option value="5">5</option>
    </param>
    <param name="cellularity" type="float" size="5" value="1" min="0" max="1" label="Cellularity" help="Ratio of tumor cells in the sample. Real value between 0 and 1"/>
	</when>
    </conditional>
    <param name="outputgraph" type="select" label="Output figures">
        <option value="TRUE">Yes</option>
        <option value="FALSE">No</option>
     </param>    
    <param name="outputlog" type="select" label="Output log">
        <option value="TRUE">Yes</option>
        <option value="FALSE">No</option>
    </param>
  </inputs>       
  <outputs>
    <data format="scr" name="output" label="segmentation of ${input.name}" />
    <data format="log" name="log" label="log of segmentation of ${input.name}">
    	<filter>outputlog == "TRUE"</filter>
    </data>
    <data format="zip" name="graph" label="graph of segmentation of ${input.name}">
    	<filter>outputgraph == "TRUE"</filter>
    </data>
  </outputs>
  <stdio>
    <exit_code range="1:"   level="fatal"   description="See logs for more details" />
   </stdio>
  <help>

**What it does**     	
This tool segments normalized profiles provided by the user and labels segments found in the copy-number profiles.
	
Input format:
  	
*A tabular text file containing 3 fixed columns and 1 column per sample:*
	
	- chr: Chromosome.
	- position: Genomic position (in bp)
  	- probeName: Probes names.
  	- One column per sample which contains the copy number profile for each sample
 
Output format:
  	
*A tabular text file containing 7 columns which describe all the segments (1 line per segment):*
	
	- sampleNames: Column names corresponding to samples in the input file.
  	- chrom: Chromosome of the segment.
	- chromStart: Starting position (in bp) of the segment. This position is not included in the segment.
	- chromEnd: Ending position (in bp) of the segment. This position is included in the segment.
	- probes: Number of probes in the segment.
	- means: Mean of the segment.
	- calls: Calling of the segment (”double loss”, ”loss”, ”normal”, ”gain” or ”amplification”).
    
-----
  		
**Citation**
If you use this tool please cite : 

`Q. Grimonprez, A. Celisse, M. Cheok, M. Figeac, and G. Marot. MPAgenomics : An R package for multi-patients analysis of genomic markers, 2014. Preprint &lt;http://fr.arxiv.org/abs/1401.5035&gt;`_
  	
If segmentation is performed with PELT, please also cite `R. Killick, P. Fearnhead, and I. A. Eckley. Optimal detection of changepoints with a linear computational cost. Journal of the American Statistical Association, 107(500):1590–1598, 2012. &lt;http://arxiv.org/abs/1101.1438&gt;`_

If segmentation is performed by cghseg, please cite	`Picard, F., Robin, S., Lavielle, M., Vaisse, C., and Daudin, J.-J. (2005). A statistical approach for array CGH data analysis. BMC Bioinformatics, 6(1):27. &lt;http://www.ncbi.nlm.nih.gov/pubmed/15705208&gt;`_ ,
and also cite Rigaill, G. (2010). `Pruned dynamic programming for optimal multiple change-point detection. &lt;http://arxiv.org/abs/1004.0887&gt;`_	

When using the labels of the segments, please cite CGHCall `M. A. van de Wiel, K. I. Kim, S. J. Vosse, W. N. van Wieringen, S. M. Wilting, and B. Ylstra. CGHcall: calling aberrations for array CGH tumor profiles. Bioinformatics, 23(7):892–894, 2007. &lt;http://bioinformatics.oxfordjournals.org/content/23/7/892.abstract&gt;`_
	
</help>
</tool>