<tool id="cg_snpdiff" name="snpdiff 1.5" version="1.0.0">
<!--
This tool creates a GUI for the snpdiff function of cgatools from Complete Genomics, Inc.
written 6-18-2012 by bcrain@completegenomics.com
-->

  <description>compares snp calls to a Complete Genomics variant file.</description> <!--adds description in toolbar-->

  <requirements>
  	<requirement type="binary">cgatools</requirement>
  </requirements>

  <command> <!--run executable-->
  	cgatools | head -1;
  	cgatools snpdiff 
  	--reference $crr.fields.path 
  	--variants $varfile
  	--genotypes $genotype
  	--output-prefix cg_
  	--reports `echo ${report1} ${report2} ${report3} | sed 's/  */,/g'` 
  </command>

  <outputs>
  	<data format="tabular" name="output1" from_work_dir="cg_Output.tsv" label="${tool.name} on ${on_string}: Output">
  	<filter>(report1 == 'Output')</filter>
  	</data>
  	<data format="tabular" name="output2" from_work_dir="cg_Verbose.tsv" label="${tool.name} on ${on_string}: Verbose">
  	<filter>(report2 == 'Verbose')</filter>
  	</data>
  	<data format="tabular" name="output3" from_work_dir="cg_Stats.tsv" label="${tool.name} on ${on_string}: Stats">
  	<filter>(report3 == 'Stats')</filter>
  	</data>
  </outputs>
  
  <inputs>
		<!--form field to select crr file-->
		<param name="crr" type="select" label="Reference genome (.crr file)">
			<options from_data_table="cg_crr_files" />
		</param>
	
		<!--conditional to select variant file input-->
  	<conditional name="data_sources">
      <param name="data_source" type="select" label="Where is the input varfile?">
        <option value="in" selected="true">imported into Galaxy</option>
        <option value="out">located outside Galaxy (available only for local Galaxy instances)</option>
      </param>
      <when value="in">
				<!--form field to select variant files-->
				<param name="varfile" type="data" format="cg_var" label="Var file">
					<validator type="unspecified_build" />
					<validator type="dataset_metadata_in_file" filename="cg_crr_files.loc"
					 metadata_name="dbkey" metadata_column="1"
					 message="cgatools is not currently available for this build."/>
				</param>
			</when>
      <when value="out">
				<!--form field to select crr file-->
				<param name="varfile" type="text" label="Variant file (/path/varfile)" size="40" help="Variant file can be compressed (gz, bz2), e.g. /harddrive/GS00000XXXX-DID/GS00000YYYY-ASM/GS00123-DNA_G01_2000/ASM/var-GS00000YYYY-ASM.tsv.bz2"/>
			</when>
		</conditional>

		<!--conditional to select genotypes file input-->
		<param name="genotype" type="data" format="tabular" label="Genotypes file with SNP calls" help="The genotypes file is a tab-delimited file with at 
    least the following columns (additional columns may be given): Chromosome (Required), Offset0Based (Required), GenotypesStrand (Optional), Genotypes (Optional)">
			<validator type="unspecified_build" />
			<validator type="dataset_metadata_in_file" filename="cg_crr_files.loc"
			 metadata_name="dbkey" metadata_column="1"
			 message="cgatools is not currently available for this build."/>
		</param>

		<param name="report1" type="select" label="Create report Output">
			<option value="">no</option>
			<option value="Output">yes</option>
		</param>
		<param name="report2" type="select" label="Create report Verbose">
			<option value="">no</option>
			<option value="Verbose">yes</option>
		</param>
		<param name="report3" type="select" label="Create report Stats">
			<option value="">no</option>
			<option value="Stats">yes</option>
		</param>

  </inputs>

  <help>
  
**What it does**

This tool ompares snp calls to a Complete Genomics variant file.

**cgatools 1.5.0 Documentation**

Userguide: http://cgatools.sourceforge.net/docs/1.5.0/cgatools-user-guide.pdf

Release notes: http://cgatools.sourceforge.net/docs/1.5.0/cgatools-release-notes.pdf

**Command line reference**::

		COMMAND NAME
		  snpdiff - Compares snp calls to a Complete Genomics variant file.
		
		DESCRIPTION
		  Compares the snp calls in the "genotypes" file to the calls in a Complete 
		  Genomics variant file. The genotypes file is a tab-delimited file with at 
		  least the following columns (additional columns may be given):
				
		    Chromosome      (Required) The name of the chromosome.
		    Offset0Based    (Required) The 0-based offset in the chromosome.
		    GenotypesStrand (Optional) The strand of the calls in the Genotypes 
		                    column (+ or -, defaults to +).
		    Genotypes       (Optional) The calls, one per allele. The following 
		                    calls are recognized:
		                    A,C,G,T A called base.
		                    N       A no-call.
		                    -       A deleted base.
		                    .       A non-snp variation.
				
		  The output is a tab-delimited file consisting of the columns of the 
		  original genotypes file, plus the following additional columns:
				
		    Reference         The reference base at the given position.
		    VariantFile       The calls made by the variant file, one per allele. 
		                      The character codes are the same as is described for 
		                      the Genotypes column.
		    DiscordantAlleles (Only if Genotypes is present) The number of 
		                      Genotypes alleles that are discordant with calls in 
		                      the VariantFile. If the VariantFile is described as 
		                      haploid at the given position but the Genotypes is 
		                      diploid, then each genotype allele is compared 
		                      against the haploid call of the VariantFile.
		    NoCallAlleles     (Only if Genotypes is present) The number of 
		                      Genotypes alleles that were no-called by the 
		                      VariantFile. If the VariantFile is described as 
		                      haploid at the given position but the Genotypes is 
		                      diploid, then a VariantFile no-call is counted twice.
				
		  The verbose output is a tab-delimited file consisting of the columns of the
		  original genotypes file, plus the following additional columns:
				
		    Reference   The reference base at the given position.
		    VariantFile The call made by the variant file for one allele (there is 
		                a line in this file for each allele). The character codes 
		                are the same as is described for the Genotypes column.
		    [CALLS]     The rest of the columns are pasted in from the VariantFile,
		                describing the variant file line used to make the call.
				
		  The stats output is a comma-separated file with several tables describing 
		  the results of the snp comparison, for each diploid genotype. The tables 
		  all describe the comparison result (column headers) versus the genotype 
		  classification (row labels) in different ways. The "Locus classification" 
		  tables have the most detailed match classifications, while the "Locus 
		  concordance" tables roll these match classifications up into "discordance" 
		  and "no-call". A locus is considered discordant if it is discordant for 
		  either allele. A locus is considered no-call if it is concordant for both 
		  alleles but has a no-call on either allele. The "Allele concordance" 
		  describes the comparison result on a per-allele basis.
		
		OPTIONS
		  -h [ --help ] 
		      Print this help message.
		
		  --reference arg
		      The input crr file.
		
		  --variants arg
		      The input variant file.
		
		  --genotypes arg
		      The input genotypes file.
		
		  --output-prefix arg
		      The path prefix for all output reports.
		
		  --reports arg (=Output,Verbose,Stats)
		      Comma-separated list of reports to generate. A report is one of:
		        Output  The output genotypes file.
		        Verbose The verbose output file.
		        Stats   The stats output file.
						
		SUPPORTED FORMAT_VERSION
		  0.3 or later
  </help>
</tool>