comparison moabs.xml @ 1:8c8cc81b34cd draft

"planemo upload for repository https://github.com/sunnyisgalaxy/moabs commit 6a45aa4c34b4f3b73ab0c6c3d9e7a315b62bf761"

0:26d7ec4af119

<tool id="moabs" name="MOABS" profile="16.04" version="@VERSION@">
    <description>MOdel based Analysis of Bisulfite Sequencing data</description>
    <macros>
        <import>macros.xml</import>
    </macros>
    <expand macro="requirements"/>

        #if str( $mcomp_advanced.doComp.compare_selector ) == "0":
            cp -f '$mcomp_advanced.doComp.compFile' comp.g1.vs.g2.txt &&
        #end if
        moabs -v 1 --def MMAP.p="\${GALAXY_SLOTS:-4}" --def MCALL.p="\${GALAXY_SLOTS:-4}" --def MCOMP.p="\${GALAXY_SLOTS:-4}" --cf '$cfg_file' &&
        #if "1" in $output_selector:
            cp -f dmr_M3_g1.G.bed_vs_g2.G.bed.txt.dmr '$output1' &&
        #end if
        #if "2" in $output_selector:
            cp -f comp.g1.vs.g2.txt '$output2' &&
        #end if
        echo Done
        ]]>
    </command>
    <configfiles>

            Path=bsmap
            d='${reference_fasta_filename}'
            #if str( $bsmap_advanced.bsmap_mismatch.bsmap_mismatch_selector ) != "0":
                v=$bsmap_advanced.bsmap_mismatch.v
            #end if
            n=$bsmap_advanced.n
            r=$bsmap_advanced.r
            R=''

            [MCALL]
            Path=mcall
            r='${reference_fasta_filename}'

            [MCOMP]
            Path=mcomp
            reference='${reference_fasta_filename}'
            doComp=$mcomp_advanced.doComp.compare_selector
        </configfile>
    </configfiles>

    <inputs>
        <conditional name="reference_source">

                    <param name="fastq_input1" format="fastqsanger,fastqsanger.gz" type="data_collection" collection_type="paired" label="Select a paired collection" help="See help section for an explanation of dataset collections"/>
                </when>
            </conditional>
        </repeat>
        <section name="bsmap_advanced" title="Advanced options for BSMAP" expanded="False">
            <conditional name="bsmap_mismatch">
                <param name="bsmap_mismatch_selector" type="select" label="Set the mismatch rate or number?" help="">
                    <option value="0">Do not set</option>
                    <option value="1">Set the mismatch rate</option>
                    <option value="2">Set the mismatch number</option>
                </param>
                <when value="1">
                    <param argument="-v" type="float" value="0.08" min="0" max="1" label="Mismatch rate" help="The mismatch rate w.r.t to the read length"/>
                </when>
                <when value="2">
                    <param argument="-v" type="integer" value="3" min="0" label="Mismatch number" help="The maximum number of mismatches allowed on a read"/>
                </when>
            </conditional>
            <param argument="-n" type="boolean" truevalue="1" falsevalue="0" checked="true" label="Mapping to four strands?" help="Yes: map SE or PE reads to all 4 strands, i.e. ++, +-, -+, --; No: only map to 2 forward strands, i.e. BSW(++) and BSC(-+)"/>
            <param argument="-r" type="select" label="How to report repeat hits" help="0=none(unique hit/pair); 1=random one; 2=all(slow)">
                <option value="0" selected="true">0</option>
                <option value="1">1</option>
                <option value="2">2</option>
            </param>
        </section>
        <section name="mcomp_advanced" title="Advanced options for MCOMP" expanded="False">
            <conditional name="doComp">
                <param name="compare_selector" type="select" label="Run the comparison or not" help="Yes: compare; No: do not compare, using the comparison result by `-c`">
                    <option value="1">Yes</option>
                    <option value="0">No</option>
                </param>
                <when value="0">
                    <param argument="-c" name="compFile" type="data" format="txt" label="Input comparison results" help="Previously generated comparison file from history"/>
                </when>
            </conditional>
        </section>
        <param name="output_selector" type="select" multiple="true" optional="true" label="Select output files" help="">
            <option value="1"> dmr_M3_g1.G.bed_vs_g2.G.bed.txt.dmr </option>
            <option value="2"> comp.g1.vs.g2.txt </option>
            <option value="3"> BAM files </option>
            <option value="4"> Methylation calling BED files </option>
        </param>
    </inputs>
    <outputs>
        <data name="output1" format="interval" label="${tool.name} on ${on_string} : dmr_M3_g1.G.bed_vs_g2.G.bed.txt.dmr">
            <filter> "1" in output_selector </filter>
        </data>
        <data name="output2" format="interval" label="${tool.name} on ${on_string} : comp.g1.vs.g2.txt">
            <filter> "2" in output_selector </filter>
        </data>
        <collection name="output_collection_bam" type="list" label="BAM files">
            <filter> "3" in output_selector </filter>
            <discover_datasets pattern="(?P&lt;designation&gt;.+\.bam$)" ext='bam'/>
        </collection>
        <collection name="output_collection_bed" type="list" label="Methylation calling BED files">
            <filter> "4" in output_selector </filter>
            <discover_datasets pattern="(?P&lt;designation&gt;g[12]\.G\.bed$)" ext='interval'/>
        </collection>
    </outputs>
    <tests>
        <test>

            <!--
            <conditional name="doComp">
                <param name="compare_selector" value="1"/>
            </conditional>
            -->
            <param name="output_selector" value="1,2,3,4"/>
            <output name="output1" file="SE_dmr_M3_g1.G.bed_vs_g2.G.bed.txt.dmr" ftype="interval" lines_diff="1"/>
            <output name="output2" file="SE_comp.g1.vs.g2.txt" ftype="interval" lines_diff="1"/>
            <output_collection name="output_collection_bam" count="4">
                <element name="g1_r1.bam" file="SE_g1_r1.bam" compare="sim_size"/>
                <element name="g1_r2.bam" file="SE_g1_r2.bam" compare="sim_size"/>
                <element name="g2_r1.bam" file="SE_g2_r1.bam" compare="sim_size"/>
                <element name="g2_r2.bam" file="SE_g2_r2.bam" compare="sim_size"/>
            </output_collection>
            <output_collection name="output_collection_bed" count="2">
                <element name="g1.G.bed" file="SE_g1.G.bed" ftype="interval" lines_diff="1"/>
                <element name="g2.G.bed" file="SE_g2.G.bed" ftype="interval" lines_diff="1"/>
            </output_collection>

            <!--
            <conditional name="doComp">
                <param name="compare_selector" value="1"/>
            </conditional>
            -->
            <param name="output_selector" value="1,2"/>
            <output name="output1" file="PE_dmr_M3_g1.G.bed_vs_g2.G.bed.txt.dmr" ftype="interval" lines_diff="1"/>
            <output name="output2" file="PE_comp.g1.vs.g2.txt" ftype="interval" lines_diff="1"/>
        </test>
        <test>
            <!-- test paired collection -->
            <param name="reference_source_selector" value="history"/>
            <param name="ref_file" ftype="fasta" value="seg.fa"/>

            <!--
            <conditional name="doComp">
                <param name="compare_selector" value="1"/>
            </conditional>
            -->
            <param name="output_selector" value="1,2"/>
            <output name="output1" file="PE_dmr_M3_g1.G.bed_vs_g2.G.bed.txt.dmr" ftype="interval" lines_diff="1"/>
            <output name="output2" file="PE_comp.g1.vs.g2.txt" ftype="interval" lines_diff="1"/>
        </test>
        <test>
            <!-- test data table reference -->
            <param name="reference_source_selector" value="cached"/>
            <param name="ref_file" value="chr11"/>

            <!--
            <conditional name="doComp">
                <param name="compare_selector" value="1"/>
            </conditional>
            -->
            <param name="output_selector" value="1,2"/>
            <output name="output1" file="SE_dmr_M3_g1.G.bed_vs_g2.G.bed.txt.dmr" ftype="interval" lines_diff="1"/>
            <output name="output2" file="SE_comp.g1.vs.g2.txt" ftype="interval" lines_diff="1"/>
        </test>
    </tests>
    <help>
        <![CDATA[
**MOABS: MOdel based Analysis of Bisulfite Sequencing data**

KO vs WT. Each group of reads may have combined sequencing library, i.e.
single-end reads and/or paired-end reads. Multiple replicates can be specified in each group.

**Outputs**

Four output files can be selected to report, namely

    1. **DMR region file** - the major result file
    2. **Comparison file between two groups** - the intermediate comparion result
    3. **BAM files** - intermediate BAM files
    4. **Methylation BED files** - intermediate methylation BED files

-----

MOABS will detect differential methylated regions (DMRs) using the input BS-Seq
reads. The output file is a tab-delimited text file (not strictly a BED
format), representing DMRs. It has 8 columns as below.

chrom<TAB>start<TAB>end<TAB>methylation_state<TAB>CpGsites<TAB>DMCcount<TAB>nonDMCcount<TAB>hidden_state

    1. **chrom** - The chromosome of the region.
    2. **start** - The start genomic locus of the region.
    3. **end** - The end genomic locus of the region.
    4. **methylation_state** - The methylation state of the region, "+"/"-" representing hyper- or hypo-methylation regions.
    5. **CpGsites** - Total number of CpG sites in the region.
    6. **DMCcount** - The number of differential methylated CpG sites (DMCs) in the region.
    7. **nonDMCcount** - The number of non-DMCs in the region.
    8. **hidden_state** - The hidden state prediced by Hidden Markov Model (HMM), "1"/"-1" representing hyper- or hypo-methylation states.

For example, six DMRs are identified in the following format.

@DMRExample@

-----

1:8c8cc81b34cd

<tool id="moabs" name="MOABS" profile="16.04" version="@VERSION@+galaxy1">
    <description>MOdel based Analysis of Bisulfite Sequencing data</description>
    <macros>
        <import>macros.xml</import>
    </macros>
    <expand macro="requirements"/>

        #if str( $mcomp_advanced.doComp.compare_selector ) == "0":
            cp -f '$mcomp_advanced.doComp.compFile' comp.g1.vs.g2.txt &&
        #end if
        moabs -v 1 --def MMAP.p="\${GALAXY_SLOTS:-4}" --def MCALL.p="\${GALAXY_SLOTS:-4}" --def MCOMP.p="\${GALAXY_SLOTS:-4}" --cf '$cfg_file' &&
        #if "1" in $output_selector:
            cp -f dmc_M2_g1.G.bed_vs_g2.G.bed.txt '$output1' &&
        #end if
        #if "2" in $output_selector:
            cp -f dmr_M2_g1.G.bed_vs_g2.G.bed.txt '$output2' &&
        #end if
        #if "3" in $output_selector:
            cp -f comp.g1.vs.g2.txt '$output3' &&
        #end if
        echo Done
        ]]>
    </command>
    <configfiles>

            Path=bsmap
            d='${reference_fasta_filename}'
            #if str( $bsmap_advanced.bsmap_mismatch.bsmap_mismatch_selector ) != "0":
                v=$bsmap_advanced.bsmap_mismatch.v
            #end if
            s=$bsmap_advanced.s
            w=$bsmap_advanced.w
            #if $bsmap_advanced.D:
                D=$bsmap_advanced.D
            #end if
            S=$bsmap_advanced.S
            n=$bsmap_advanced.n
            q=$bsmap_advanced.q
            z=$bsmap_advanced.z
            f=$bsmap_advanced.f
            #if $bsmap_advanced.A:
                A=$bsmap_advanced.A
            #end if
            r=$bsmap_advanced.r
            #if str( $bsmap_advanced.R ) == "1":
                R=''
            #end if
            #if str( $bsmap_advanced.u ) == "1":
                u=''
            #end if
            m=$bsmap_advanced.m
            x=$bsmap_advanced.x

            [MCALL]
            Path=mcall
            r='${reference_fasta_filename}'
            cytosineMinScore=$mcall_advanced.cytosineMinScore
            nextBaseMinScore=$mcall_advanced.nextBaseMinScore
            qualityScoreBase=$mcall_advanced.qualityScoreBase
            trimWGBSEndRepairPE2Seq=$mcall_advanced.trimWGBSEndRepairPE2Seq
            trimWGBSEndRepairPE1Seq=$mcall_advanced.trimWGBSEndRepairPE1Seq
            processPEOverlapSeq=$mcall_advanced.processPEOverlapSeq
            trimRRBSEndRepairSeq=$mcall_advanced.trimRRBSEndRepairSeq
            minFragSize=$mcall_advanced.minFragSize
            minMMFragSize=$mcall_advanced.minMMFragSize
            reportCpX=$mcall_advanced.reportCpX

            [MCOMP]
            Path=mcomp
            doComp=$mcomp_advanced.doComp.compare_selector
            d=$mcomp_advanced.d
            filterCredibleDif=$mcomp_advanced.filterCredibleDif
            pFetDmc=$mcomp_advanced.pFetDmc
            pFetDmr=$mcomp_advanced.pFetDmr
            minNominalDif=$mcomp_advanced.minNominalDif
            minCredibleDif=$mcomp_advanced.minCredibleDif
            minDmcsInDmr=$mcomp_advanced.minDmcsInDmr
            maxDistConsDmcs=$mcomp_advanced.maxDistConsDmcs
        </configfile>
    </configfiles>

    <inputs>
        <conditional name="reference_source">

                    <param name="fastq_input1" format="fastqsanger,fastqsanger.gz" type="data_collection" collection_type="paired" label="Select a paired collection" help="See help section for an explanation of dataset collections"/>
                </when>
            </conditional>
        </repeat>
        <section name="bsmap_advanced" title="Advanced options for BSMAP" expanded="False">
            <param argument="-s" type="integer" value="16" min="8" max="16" label="Seed size" help="The seed size for the HASH table. BSMAP implements a seeding algorithm by indexing reference for all possible k-mers, i.e. seeds. As for the seed size, i.e. the length of k-mers, 16 is suggested for the WGBS mode, and 12 is suggested for the RRBS mode. Min=8, max=16."/>
            <conditional name="bsmap_mismatch">
                <param name="bsmap_mismatch_selector" type="select" label="Set the mismatch rate or number?" help="When `Do not set` selected, BSMAP will allow a default suggested 8% mismatch rate. Otherwise, a customized mismatch can be controlled by specifying a mismatch rate or a mismatch number.">
                    <option value="0" selected="true">Do not set</option>
                    <option value="1">Set the mismatch rate</option>
                    <option value="2">Set the mismatch number</option>
                </param>
                <when value="1">
                    <param argument="-v" type="float" value="0.08" min="0" max="1" label="Mismatch rate" help="The mismatch rate w.r.t to the read length"/>
                </when>
                <when value="2">
                    <param argument="-v" type="integer" value="3" min="0" label="Mismatch number" help="The maximum number of mismatches allowed on a read"/>
                </when>
            </conditional>
            <param argument="-w" type="integer" value="1000" min="0" max="1000" label="Maximum number of equal best hits to count" help="Maximum number of equal best hits to count. When multiple mapping occurs for a read, it should control the number of records to report. Default: 1000."/>
            <param argument="-D" type="text" label="Restriction enzyme digestion sites for RRBS mode" help="For RRBS data, this option activates RRBS mapping mode and set restriction enzyme digestion sites. Digestion position marked by '-', example: -D C-CGG for MspI digestion.  default: none (WGBS mode).">
              <validator type="regex" message="Use A/T/C/G/- for restriction enzyme digestion sites">^[ATCG-]*$</validator>
            </param>
            <param argument="-S" type="integer" value="0" label="Random seed" help="Seed for random number generation used in selecting multiple hits. Other seed values generate pseudo random number based on read index number, to allow reproducible mapping results. Default=0 (get seed from system clock, mapping results not resproducible)."/>
            <param argument="-n" type="boolean" truevalue="1" falsevalue="0" checked="true" label="Mapping to four strands?" help="Yes: map SE or PE reads to all 4 strands, i.e. ++, +-, -+, --; No: only map to 2 forward strands, i.e. BSW(++) and BSC(-+). For example, for a traditional library construction, two forward strands, ++ and -+, are sufficient for alignments. However, a Pico library construction needs all four-strand mappings."/>
            <param argument="-q" type="integer" value="0" min="0" max="40" label="Quality threshold in trimming" help="The quality threshold to trim read bases. To obtain an accurate mapping, low-quality bases should be skipped beforehand. Min=0, max=40. Default=0 (no trim)."/>
            <param argument="-z" type="select" label="Base quality" help="Base quality for sequencing reads, Illumina or Sanger.">
                <option value="33" selected="true">Sanger</option>
                <option value="64">Illumina</option>
            </param>
            <param argument="-f" type="integer" value="5" min="0" label="Maximum number of Ns in a read to filter out" help="To filter out low-quality reads containing >n Ns. Default=5."/>
            <param argument="-A" type="text" label="3' adapter sequence to trim" help="To trim 3' adapter sequence. Default: none (no trim).">
              <validator type="regex" message="Use A/T/C/G for adapter sequences, and the length should be greater equal to 12 bases.">^[ATCG]{12,}$|^$</validator>
            </param>
            <param argument="-r" type="select" label="How to report repeat hits" help="0=none(unique hit/pair); 1=random one; 2=all(slow). When input reads coverage is high, it is suggested to report only unique hits (r=0). For a low-depth library, a random one from multiple mappings (r=1), or all multiple mappings (r=2) can be specified to increase read coverage, yet be cautious about bias caused by ambiguous mappings.">
                <option value="0" selected="true">0</option>
                <option value="1">1</option>
                <option value="2">2</option>
            </param>
            <param argument="-R" type="boolean" truevalue="1" falsevalue="0" checked="false" label="Print corresponding reference sequences?" help="Yes: print corresponding reference sequences in mapping records, a `RS:` tag will be added in record attributes; No: do not print reference sequences."/>
            <param argument="-u" type="boolean" truevalue="1" falsevalue="0" checked="false" label="Report unmapped reads?" help="Yes: print unmapped reads; No: do not print unmapped reads."/>
            <param argument="-m" type="integer" value="28" min="0" label="Minimal insert size allowed in paired-end mapping" help="For paired-end reads, the minimal insert size allowed in two end mapping. Default=28."/>
            <param argument="-x" type="integer" value="1000" min="0" label="Maximal insert size allowed in paired-end mapping" help="For paired-end reads, the maximal insert size allowed in two end mapping. Default=1000."/>
        </section>
        <section name="mcall_advanced" title="Advanced options for MCALL" expanded="False">
            <param argument="--cytosineMinScore" type="integer" value="20" min="0" label="Threshold for cytosine quality score" help="Threshold for cytosine quality score. Discard the base if threshold is not reached. Default=20."/>
            <param argument="--nextBaseMinScore" type="integer" value="3" min="-1" label="Threshold for the next base quality score" help="Threshold for the next base quality score. Possible values: -1 makes the program not to check if next base matches reference; any positive integer or zero makes the program to check if next base matches reference and reaches this score threshold; default=3, i.e., better than 'B' or '#'."/>
            <param argument="--qualityScoreBase" type="select" label="Specify the quality score system" help="Sanger, Solexa, or Illumina. See wiki FASTQ_format for details. Default: auto-detection.">
                <option value="0" selected="true">Auto-detection</option>
                <option value="33">Sanger</option>
                <option value="59">Solexa</option>
                <option value="64">Illumina</option>
            </param>
            <param argument="--trimWGBSEndRepairPE2Seq" type="integer" value="3" min="0" label="Bases to trim end-repair sequences from +-/--" help="To trim end-repair sequence from begin of +-/-- reads from Pair End WGBS Sequencing. 0: no trim; n (positive integer): trim n bases from begin of +-/-- reads. Default: trim 3 bases."/>
            <param argument="--trimWGBSEndRepairPE1Seq" type="integer" value="3" min="0" label="Bases to trim end-repair sequences from ++/-+" help="To trim end-repair sequence from end of ++/-+ reads from Pair End WGBS Sequencing. 0: no trim; n (positive integer): trim n bases from end of ++/-+ reads. Default: trim 3 bases."/>
            <param argument="--processPEOverlapSeq" type="select" label="Count once or twice the overlap sequence of two pairs" help="Two ends of paired-end reads may be overlapped in mapping. The overlap sequencce will be counted once or twice for cytosine methylation measurements. Default: once.">
                <option value="1" selected="true">Once</option>
                <option value="0">Twice</option>
            </param>
            <param argument="--trimRRBSEndRepairSeq" type="select" label="How to trim end-repair sequence for RRBS reads?" help="To trim end-repair sequence for RRBS reads. 0: no trim; 2: trim the last CG at exactly end of ++/-+ reads and trim the first CG at exactly begin of +-/-- reads like the WGBS situation. Default=2.">
                <option value="2" selected="true">2</option>
                <option value="0">0</option>
            </param>
            <param argument="--minFragSize" type="integer" value="0" min="0" label="Minimal fragment size for properly mapped reads" help="To retain properly mapped and large enough fragment sizes. The 9th field in the BAM file is the fragment size of the mapping, and non-properly-paired read has 0 at the 9th field. This option is set to require properly paired and large enough fragment size. Default=0 for all records."/>
            <param argument="--minMMFragSize" type="integer" value="0" min="0" label="Minimal fragment size for multiply matched read" help="Same as --minFragSize but this option is only applicable to reads with flag 0x100 set as 1, i.e., reads multiply mapped. Default=0 for all records."/>
            <param argument="--reportCpX" type="select" label="Generates CpG/A/C/T methylation?" help="To generate methylation for CpG, or CpA/CpC/CpT. Default=CpG.">
                <option value="G" selected="true">CpG</option>
                <option value="C">CpC</option>
                <option value="A">CpA</option>
                <option value="T">CpT</option>
            </param>
        </section>
        <section name="mcomp_advanced" title="Advanced options for MCOMP" expanded="False">
            <conditional name="doComp">
                <param name="compare_selector" type="select" label="Run the comparison or not" help="Yes: compare; No: do not compare, using the comparison result by `-c`">
                    <option value="1" selected="true">Yes</option>
                    <option value="0">No</option>
                </param>
                <when value="0">
                    <param argument="-c" name="compFile" type="data" format="txt" label="Input comparison results" help="Previously generated comparison file from history"/>
                </when>
            </conditional>
            <param argument="-d" type="integer" value="3" min="0" label="Minimum depth for a site coverage" help="If a site has depth less than `d`, this site is ignored for statistical tests. This option affects much of nominal ratios but none of credible ratios. This option may be reset during later DMC/DMR rescan to filter sites with depth less than `d`. Default=3."/>
            <param argument="--filterCredibleDif" type="float" value="-10" label="Minimum absolute credible methylation difference (CDIF)" help="If absolute value of CDIF for a site less than filterCredibleDif, this site is ignored for regional calculation. Use a small value, such as 0.01, to filter all sites with no difference; use 0.20 (for example) to select DMCs. Any negative number means no filter. Default=-10."/>
            <param argument="--pFetDmc" type="float" value="0.05" min="0" max="1" label="Cutoff of Pvalue from Fisher Exact Test for DMC scan" help="Cutoff of P value from Fisher Exact Test for DMC scan. Default=`0.05`."/>
            <param argument="--pFetDmr" type="float" value="0.05" min="0" max="1" label="Cutoff of Pvalue from Fisher Exact Test for DMR scan" help="Cutoff of P value from Fisher Exact Test for DMR scan. Default=`0.05`."/>
            <param argument="--minNominalDif" type="float" value="0.33333" min="0" max="1" label="Minimal nominal methylation difference for DMC and DMR calling" help="Minimal nominal methylation difference for DMC and DMR. Default=`0.3333`."/>
            <param argument="--minCredibleDif" type="float" value="0.2" min="0" max="1" label="Minimal credible methylation difference for DMC calling" help="Minimal CDIF for DMC calling. Default=`0.2`."/>
            <param argument="--minDmcsInDmr" type="integer" value="3" min="1" label="Minimum number of DMCs in a DMR" help="Minimum number of DMCs in a DMR. Default=3."/>
            <param argument="--maxDistConsDmcs" type="integer" value="300" min="1" label="Maximum distance between two consective DMCs for a DMR" help="Maximum distance between two consective DMCs for a DMR. Default=300."/>
        </section>
        <param name="output_selector" type="select" multiple="true" optional="true" label="Select output files" help="">
            <option value="1" selected="true">dmc_M2_g1.G.bed_vs_g2.G.bed.txt</option>
            <option value="2" selected="true">dmr_M2_g1.G.bed_vs_g2.G.bed.txt</option>
            <option value="3" selected="true">comp.g1.vs.g2.txt</option>
            <option value="4" selected="true">BAM files</option>
            <option value="5" selected="true">Methylation calling BED files</option>
        </param>
    </inputs>
    <outputs>
        <data name="output1" format="interval" label="${tool.name} on ${on_string} : dmc_M2_g1.G.bed_vs_g2.G.bed.txt">
            <filter> "1" in output_selector </filter>
        </data>
        <data name="output2" format="interval" label="${tool.name} on ${on_string} : dmr_M2_g1.G.bed_vs_g2.G.bed.txt">
            <filter> "2" in output_selector </filter>
        </data>
        <data name="output3" format="interval" label="${tool.name} on ${on_string} : comp.g1.vs.g2.txt">
            <filter> "3" in output_selector </filter>
        </data>
        <collection name="output_collection_bam" type="list" label="BAM files">
            <filter> "4" in output_selector </filter>
            <discover_datasets pattern="(?P&lt;designation&gt;.+\.bam$)" ext='bam'/>
        </collection>
        <collection name="output_collection_bed" type="list" label="Methylation calling BED files">
            <filter> "5" in output_selector </filter>
            <discover_datasets pattern="(?P&lt;designation&gt;g[12]\.G\.bed$)" ext='interval'/>
        </collection>
    </outputs>
    <tests>
        <test>

            <!--
            <conditional name="doComp">
                <param name="compare_selector" value="1"/>
            </conditional>
            -->
            <param name="output_selector" value="1,2,3,4,5"/>
            <output name="output1" file="SE_dmc_M2_g1.G.bed_vs_g2.G.bed.txt" ftype="interval" lines_diff="1"/>
            <output name="output2" file="SE_dmr_M2_g1.G.bed_vs_g2.G.bed.txt" ftype="interval" lines_diff="1"/>
            <output name="output3" file="SE_comp.g1.vs.g2.txt" ftype="interval" lines_diff="1"/>
            <output_collection name="output_collection_bam" count="4">
                <element name="g1_r1.bam" file="SE_g1_r1.bam" ftype="bam" lines_diff="2"/>
                <element name="g1_r2.bam" file="SE_g1_r2.bam" ftype="bam" lines_diff="2"/>
                <element name="g2_r1.bam" file="SE_g2_r1.bam" ftype="bam" lines_diff="2"/>
                <element name="g2_r2.bam" file="SE_g2_r2.bam" ftype="bam" lines_diff="2"/>
            </output_collection>
            <output_collection name="output_collection_bed" count="2">
                <element name="g1.G.bed" file="SE_g1.G.bed" ftype="interval" lines_diff="1"/>
                <element name="g2.G.bed" file="SE_g2.G.bed" ftype="interval" lines_diff="1"/>
            </output_collection>

            <!--
            <conditional name="doComp">
                <param name="compare_selector" value="1"/>
            </conditional>
            -->
            <param name="output_selector" value="1,2,3"/>
            <output name="output1" file="PE_dmc_M2_g1.G.bed_vs_g2.G.bed.txt" ftype="interval" lines_diff="1"/>
            <output name="output2" file="PE_dmr_M2_g1.G.bed_vs_g2.G.bed.txt" ftype="interval" lines_diff="1"/>
            <output name="output3" file="PE_comp.g1.vs.g2.txt" ftype="interval" lines_diff="1"/>
        </test>
        <test>
            <!-- test paired collection -->
            <param name="reference_source_selector" value="history"/>
            <param name="ref_file" ftype="fasta" value="seg.fa"/>

            <!--
            <conditional name="doComp">
                <param name="compare_selector" value="1"/>
            </conditional>
            -->
            <param name="output_selector" value="1,2,3"/>
            <output name="output1" file="PE_dmc_M2_g1.G.bed_vs_g2.G.bed.txt" ftype="interval" lines_diff="1"/>
            <output name="output2" file="PE_dmr_M2_g1.G.bed_vs_g2.G.bed.txt" ftype="interval" lines_diff="1"/>
            <output name="output3" file="PE_comp.g1.vs.g2.txt" ftype="interval" lines_diff="1"/>
        </test>
        <test>
            <!-- test data table reference -->
            <param name="reference_source_selector" value="cached"/>
            <param name="ref_file" value="chr11"/>

            <!--
            <conditional name="doComp">
                <param name="compare_selector" value="1"/>
            </conditional>
            -->
            <param name="output_selector" value="1,2,3"/>
            <output name="output1" file="SE_dmc_M2_g1.G.bed_vs_g2.G.bed.txt" ftype="interval" lines_diff="1"/>
            <output name="output2" file="SE_dmr_M2_g1.G.bed_vs_g2.G.bed.txt" ftype="interval" lines_diff="1"/>
            <output name="output3" file="SE_comp.g1.vs.g2.txt" ftype="interval" lines_diff="1"/>
        </test>
    </tests>
    <help>
        <![CDATA[
**MOABS: MOdel based Analysis of Bisulfite Sequencing data**

KO vs WT. Each group of reads may have combined sequencing library, i.e.
single-end reads and/or paired-end reads. Multiple replicates can be specified in each group.

**Outputs**

Five output files can be selected to report, namely

    1. **DMC site file** - the major DMC result file
    2. **DMR region file** - the major DMR result file
    3. **Comparison file between two groups** - the intermediate comparion result
    4. **BAM files** - intermediate BAM files
    5. **Methylation BED files** - intermediate methylation BED files

-----

MOABS detects differentially methylated cytosines (DMCs) and differentially
methylated regions (DMRs) using the input BS-Seq reads. The output DMC and DMR
file are tab-delimited text files (not strictly a BED format), representing
DMCs and DMRs.

A DMC site file has 15 columns as below.

chrom<TAB>start<TAB>end<TAB>totalC_0<TAB>nominalRatio_0<TAB>ratioCI_0<TAB>totalC_1<TAB>nominalRatio_1<TAB>ratioCI_1<TAB>nominalDif_1-0<TAB>credibleDif_1-0<TAB>difCI_1-0<TAB>p_sim_1_v_0<TAB>p_fet_1_v_0<TAB>class

    1. **chrom** - The chromosome of the CpG site.
    2. **start** - The start genomic locus of the CpG site.
    3. **end** - The end genomic locus of the CpG site.
    4. **totalC_0** - The total number of CpG read coverage in group 0.
    5. **nominalRatio_0** - The nominal methylation ratio of the CpG in group 0.
    6. **ratioCI_0** - The confidence interval (CI) of the nominal methylation ratio at the CpG site in group 0.
    7. **totalC_1** - The total number of CpG read coverage in group 1.
    8. **nominalRatio_1** - The nominal methylation ratio of the CpG in group 1.
    9. **ratioCI_1** - The confidence interval (CI) of the nominal methylation ratio at the CpG site in group 1.
    10. **nominalDif_1-0** - The nominal methylation difference between the group 1 and the group 0.
    11. **credibleDif_1-0** - The credible methylation difference (CDIF) between the group 1 and the group 0.
    12. **difCI_1-0** - The difference of ratio CIs between the group 1 and the group 0.
    13. **p_sim_1_v_0** - P-value according to the similarity probablities.
    14. **p_fet_1_v_0** - P-value according to the Fisher exact test.
    15. **class** - 5-state class labels by methylation differences and p-values.

For example, CpGs in the DMC file are recorded in the following format.

@DMCExample@

A DMR result file has 12 columns as below.

chrom<TAB>start<TAB>end<TAB>meanRatio_0<TAB>totalC_0<TAB>cSites_0<TAB>meanRatio_1<TAB>totalC_1<TAB>cSites_1<TAB>methDif_1-0<TAB>p_1_v_0<TAB>class_1_v_0

    1. **chrom** - The chromosome of the region.
    2. **start** - The start genomic locus of the region.
    3. **end** - The end genomic locus of the region.
    4. **meanRatio_0** - Mean methylation ratio of the region in group 0.
    5. **totalC_0** - Total cytosine coverage of the region in group 0.
    6. **cSites_0** - The number of CpG sites of the region in group 0.
    7. **meanRatio_1** - Mean methylation ratio of the region in group 1.
    8. **totalC_1** - Total cytosine coverage of the region in group 1.
    9. **cSites_1** - The number of CpG sites of the region in group 1.
    10. **methDif_1-0** - Average methylation difference of the region between group 1 and group 0.
    11. **p_1_v_0** - P-value from Fisher exact test of the region between group 1 and group 0.
    12. **class_1_v_0** - 5-state class labels for the DMR.

For example, four DMRs are identified in the following format.

@DMRExample@

-----