<tool id="bismark_deduplicate" name="Bismark Deduplicate" version="0.16.3">

    <description>Deduplicates reads mapped by Bismark</description>
    <!--<version_command>bismark version</version_command>-->

    <requirements>
        <requirement type="package" version="0.1.19">samtools</requirement>
        <requirement type="package" version="2.1.0">bowtie2</requirement>
    </requirements>

    <stdio>
        <exit_code range="1:" />
        <exit_code range=":-1" />
        <regex match="Error:" />
        <regex match="Exception:" />
    </stdio>

    <command interpreter="python">
<![CDATA[
        bismark_deduplicate_wrapper.py

        --tool_dir "$__tool_directory__"

        #if str ( $sPaired ) == "single":
        	-s
        #else
        	-p
        #end if

        --input "$mapping_output"

        --output_report "$output_report"
        --output_bam "$output_bam"

        ##--log_report "$log_report"
]]>
    </command>

    <inputs>
        <param name="sPaired" type="select" label="Is this library mate-paired?" format="bam">
            <option value="single">Single-end</option>
            <option value="paired">Paired-end</option>
        </param>
        <param name="mapping_output" type="data" format="bam, sam" label="Submit the resulting bam/sam file from Bismark bisulfite mapper" />
    </inputs>

    <outputs>
        <data name="output_bam" format="bam" label="${tool.name} on ${on_string}: deduplicated mapped reads" />
        <data name="output_report" format="txt" label="${tool.name} on ${on_string}: deduplication report"/>
        <!--<data name="log_report" format="txt" label="${tool.name} on ${on_string}: log report (tool stdout)"/>-->
    </outputs>

    <help>
<![CDATA[
**What it does**

	| This tool is supposed to remove alignments to the same position in the genome from the Bismark mapping output (both single and paired-end SAM files), which can arise by e.g. excessive PCR amplification. If sequences align to the same genomic position but on different strands they will be scored individually.
	|
	| Note that deduplication is not recommended for RRBS-type experiments!
	|
	| For single-end alignments only use the start-coordinate of a read will be used for deduplication.
	| For paired-end alignments the start-coordinate of the first read and the end coordinate of the second read will be used for deduplication. 

]]>
  </help>
  <citations>
      <citation type="doi">10.1093/bioinformatics/btr167</citation>
  </citations>
</tool>