annotate macs21_wrapper.xml @ 3:4124781932db draft

Updated to MACS 2.1.1 and use conda for dependency resolution.

<tool id="macs2_1_peakcalling" name="MACS2.1.1" version="2.1.1.0">
  <description>Model-based Analysis of ChIP-Seq: peak calling</description>
  <requirements>
    <requirement type="package" version="2.7">python</requirement>
    <requirement type="package" version="2.1.1">macs2</requirement>
    <requirement type="package" version="3.4">R</requirement>
    <requirement type="package" version="357">ucsc-fetchchromsizes</requirement>
    <requirement type="package" version="357">ucsc-bedclip</requirement>
    <requirement type="package" version="357">ucsc-bedsort</requirement>
    <requirement type="package" version="357">ucsc-bedgraphtobigwig</requirement>
  </requirements>
  <version_command>macs2 --version</version_command>
  <command><![CDATA[
    python $__tool_directory__/macs21_wrapper.py callpeak
    ##
    ## ChIP-seq input
    $input_chipseq_file1
    ##
    ## ChIP-seq control
    #if str($input_control_file1) != 'None'
       -c $input_control_file1
    #end if
    ##
    --format=$format
    --name="$experiment_name"
    --bw=$bw
    ##
    ## Genome size
    #if str($genome_size.gsize) == ''
       --gsize=$genome_size.user_defined_gsize
    #else:
       --gsize=$genome_size.gsize
    #end if
    ##
    ## Broad peaks
    #if str($broad_options.broad_regions) == 'broad'
       --broad --broad-cutoff=$broad_options.broad_cutoff
    #end if
    ##
    ## (no)model options
    #if str($nomodel_type.nomodel_type_selector) == 'nomodel'
       --nomodel --extsize=$nomodel_type.extsize
    #end if
    ##
    ## pq value select options
    #if str($pq_options.pq_options_selector) == 'qvalue'
       --qvalue=$pq_options.qvalue
    #else
       --pvalue=$pq_options.pvalue
    #end if
    ##
    ## Bedgraph options
    #if $bdg_options.bdg
       -B $bdg_options.spmr
    #end if
    ##
    ## Advanced options
    #if $advanced_options.advanced_options_selector
       --mfold $advanced_options.mfoldlo $advanced_options.mfoldhi
       $advanced_options.nolambda
       $advanced_options.call_summits
       #if str($advanced_options.keep_duplicates.keep_dup) == ''
          --keep-dup $advanced_options.keep_duplicates.maximum_tags
       #else
          --keep-dup $advanced_options.keep_duplicates.keep_dup
       #end if
    #else
       ## Defaults if advanced options not set
       --mfold 10 30 --keep-dup 1
    #end if
    ##
    ## Output files
    --output-summits=$output_summits_bed_file
    --output-extra-files=$output_extra_files
    --output-extra-files-path=$output_extra_files.files_path
    ##
    ## Narrow/broad peak outputs
    #if str($broad_options.broad_regions) == 'broad'
       --output-broadpeaks=$output_broadpeaks_file
       --output-gappedpeaks=$output_gappedpeaks_file
    #else
       --output-narrowpeaks=$output_narrowpeaks_file
    #end if
    ##
    ## Bedgraph outputs
    #if $bdg_options.bdg
       --output-pileup=$output_treat_pileup_file 
       --output-lambda-bedgraph=$output_lambda_bedgraph_file
       #if $bdg_options.make_bigwig
          --output-bigwig=$output_bigwig_file
          --length=$GALAXY_DATA_INDEX_DIR/shared/ucsc/chrom/${input_chipseq_file1.dbkey}.len
       #end if
    #end if
    ##
    ## XLS/interval output
    #if str($xls_to_interval) == 'True'
       --output-xls-to-interval=$output_xls_to_interval_peaks_file
    #else
       --output-peaks=$output_peaks_file
    #end if
  ]]></command>
  <inputs>
    <!--experiment name used as base for output file names -->
    <param name="experiment_name" type="text" value="MACS2.1 in Galaxy" size="50"
	   label="Experiment Name"/>
    <!--choose 'broad' or 'narrow' regions-->
    <conditional name="broad_options">
      <param name="broad_regions" type="select" label="Type of region to call"
	     help="Broad regions are formed by linking nearby enriched regions">
	<option value="" selected="true">Narrow regions</option>
	<option value="broad">Broad regions</option>
      </param>
      <when value="broad">
	<param name="broad_cutoff" type="float"
	       label="Cutoff for broad regions"
	       value="0.1" help="default: 0.1 (--broad-cutoff)"/>
      </when>
    </conditional>
    <param name="format" type="select" label="Format of input read data"
	   help="Specify the format of the input data and whether or not it is paired end (--format)">
      <option value="BAMPE" selected="true">BAM (paired-end)</option>
      <option value="BAM">BAM (single-end)</option>
      <option value="BEDPE">BED (paired-end)</option>
      <option value="BED">BED (single-end)</option>
      <option value="SAMPE">SAM (paired-end)</option>
      <option value="SAM">SAM (single-end)</option>
    </param>
    <param name="input_chipseq_file1" type="data" format="bed,sam,bam"
	   label="ChIP-seq read file" />
    <param name="input_control_file1" type="data" format="bed,sam,bam" optional="True"
	   label="ChIP-seq control read file" />
    <conditional name="genome_size">
      <param name="gsize" type="select" label="Effective genome size"
	     help="Either pre-defined (for common organisms), or user-defined (--gsize)">
	<option value="hs" selected="true">Human (2.7e9)</option>
	<option value="mm">Mouse (1.87e9)</option>
	<option value="ce">C. elegans (9e7)</option>
	<option value="dm">Fruitfly (1.2e8)</option>
	<option value="">User-defined</option>
      </param>
      <when value="">
	<!-- User-defined effective genome size -->
	<param name="user_defined_gsize" type="float" value=""
	       label="Enter effective genome size (number of bases)"
	       help="e.g. '1.0e+9' or '1000000000'" />
      </when>
    </conditional>
    <param name="bw" type="integer" label="Band width" value="300" help="(--bw)"/>
    <param name="xls_to_interval" label="Include XLS file from MACS"
	   type="boolean" truevalue="True" falsevalue="False" checked="True"
	   help="MACS2 XLS file will be output to the history in 'interval' format (suitable for subsequent analysis in Galaxy). Note that start positions are 1-based."/>

    <conditional name="bdg_options">
      <param name="bdg"
	     label="Save treatment and control lambda pileups in bedGraph"
	     type="boolean" truevalue="-B" falsevalue="" checked="False" />
      <when value="-B">
	<param name="spmr"
	       type="boolean" truevalue="--SPMR" falsevalue="" checked="False"
	       label="Save signal per million reads for fragment pileup profiles"
	       help="(--SPMR)" />
	<param name="make_bigwig" type="boolean" checked="True"
	       truevalue="True" falsevalue=""
	       label="Also generate bigWig file from bedGraph"
	       help="bigWig file can used in subsequent analyses e.g. CEAS" />
      </when>
      <when value="">
	<!-- Display nothing -->
      </when>
    </conditional>

    <conditional name="pq_options">
      <param name="pq_options_selector" type="select"
	     label="Select p-value or q-value" help="default uses q-value">
	<option value="qvalue">q-value</option>
	<option value="pvalue">p-value</option>
      </param>
      <when value="pvalue">
	<param name="pvalue" type="float"
	       label="p-value cutoff for binding region detection"
	       value="1e-2" help="default: 1e-2 (--pvalue)"/>
      </when>
      <when value="qvalue">
	<param name="qvalue" type="float"
	       label="q-value cutoff for binding region detection"
	       value="0.01" help="default: 0.01 (--qvalue)"/>
      </when>
    </conditional>
    <conditional name="advanced_options">
      <param name="advanced_options_selector"
	     type="boolean" truevalue="on" falsevalue="off" checked="False"
	     label="Use advanced options?" />
      <when value="on">
        <param name="mfoldlo" type="integer"
	       label="Select the regions with MFOLD high-confidence enrichment ratio against background to build model (lower-limit)"
	       value="10" help="(--mfold)"/>
	<param name="mfoldhi" type="integer"
	       label="Select the regions with MFOLD high-confidence enrichment ratio against background to build model (upper-limit)"
	       value="30" help="(--mfold)"/>
	<param name="nolambda"
	       label="Use fixed background lambda as local lambda for every binding region"
	       type="boolean" truevalue="--nolambda" falsevalue="" checked="False"
	       help="(--nolambda)"/>
	<param name="call_summits"
	       label="Detect subpeaks within binding region"
	       type="boolean" truevalue="--call-summits" falsevalue="" checked="False"
	       help="(--call-summits)"/>
	<conditional name="keep_duplicates">
	  <param name="keep_dup" type="select"
		 label="Use of duplicate reads">
	    <option value="auto">Automatically calculate maximum number of duplicates to keep (auto)</option>
	    <option value="all">Use all duplicates (all)</option>
	    <option value="" selected="true">Manually specify maxium number of duplicates</option>
	  </param>
	  <when value="">
	    <param name="maximum_tags" type="integer" value="1"
		   label="Maxium number of duplicated tags to keep at each location"/>
	  </when>
	</conditional>
      </when>
      <when value="off">
	<!--display nothing-->
      </when>
    </conditional>
    <conditional name="nomodel_type">
      <param name="nomodel_type_selector" type="select" label="Build Model">
	<option value="nomodel">Do not build the shifting model (--nomodel enabled)</option>
        <option value="create_model" selected="true">Build the shifting model (--nomodel disabled)</option>
      </param>
      <when value="nomodel">
        <param name="extsize" type="integer" label="Arbitrary extension size in bp" value="200" help="Used as fragment size to extend each read towards 3' end (--extsize)"/>
      </when>
    </conditional>
  </inputs>

  <outputs>
    <!--callpeaks output-->
    <data name="output_extra_files" format="html"
	  label="${tool.name}: callpeak on ${on_string} (html report)">
    </data>
    <data name="output_summits_bed_file" format="bed"
	  label="${tool.name}: callpeak on ${on_string} (summits: bed)">
    </data>
    <data name="output_peaks_file" format="xls"
	  label="${tool.name}: callpeak on ${on_string} (peaks: xls)">
      <filter>xls_to_interval is False</filter>
    </data>
    <data name="output_narrowpeaks_file" format="interval"
	  label="${tool.name}: callpeak on ${on_string} (peaks: narrowPeak)">
      <filter>broad_options['broad_regions'] == ''</filter>
    </data>
    <data name="output_broadpeaks_file" format="interval"
	  label="${tool.name}: callpeak on ${on_string} (peaks: broadPeak)">
      <filter>broad_options['broad_regions'] == 'broad'</filter>
    </data>
    <data name="output_gappedpeaks_file" format="interval"
	  label="${tool.name}: callpeak on ${on_string} (peaks: gappedPeak)">
      <filter>broad_options['broad_regions'] == 'broad'</filter>
    </data>
    <data name="output_xls_to_interval_peaks_file" format="interval"
	  label="${tool.name}: callpeak on ${on_string} (peaks: interval)">
      <filter>xls_to_interval is True</filter>
    </data>
    <data name="output_treat_pileup_file" format="bedgraph"
	  label="${tool.name}: callpeak on ${on_string} (treat pileup: bedGraph)">
      <filter>bdg_options['bdg'] is True</filter>
    </data>
    <data name="output_lambda_bedgraph_file" format="bedgraph"
	  label="${tool.name}: callpeak on ${on_string} (control lambda: bedGraph)">
      <filter>bdg_options['bdg'] is True</filter>
    </data>
    <data name="output_bigwig_file" format="bigwig"
	  label="${tool.name}: callpeak on ${on_string} (treat pileup: bigWig)">
      <filter>bdg_options['bdg'] is True</filter>
      <filter>bdg_options['make_bigwig'] is True</filter>
    </data>
  </outputs>
  <tests>
    <!-- Peak calling without bigwig output -->
    <test>
      <!-- Inputs -->
      <param name="experiment_name" value="test_MACS2.1.1" />
      <param name="broad_regions" value="" />
      <param name="format" value="BED" />
      <param name="input_chipseq_file1" value="test_region_IP.bed" dbkey="galGal3"
	     ftype="bed" />
      <param name="input_control_file1" value="test_region_Input.bed"
	     ftype="bed" />
      <param name="gsize" value="" />
      <param name="user_defined_gsize" value="775000000.0" />
      <param name="bw" value="300" />
      <param name="xls_to_interval" value="true" />
      <param name="bdg_options|bdg" value="-B" />
      <param name="bdg_options|spmr" value="--SPMR" />
      <param name="bdg_options|make_bigwig" value="false" />
      <param name="pq_options_selector" value="qvalue" />
      <param name="qvalue" value="0.05" />
      <param name="advanced_options_selector" value="true" />
      <param name="advanced_options|mfoldlo" value="5" />
      <param name="advanced_options|mfoldhi" value="50" />
      <param name="advanced_options|nolambda" value="" />
      <param name="advanced_options|call_summits" value="" />
      <param name="advanced_options|keep_duplicates" value="" />
      <param name="advanced_options|maximum_tags" value="1" />
      <param name="nomodel_type_selector" value="nomodel" />
      <param name="nomodel_type|extsize" value="243" />
      <!-- Outputs -->
      <output name="output_extra_files" file="test_MACS2.1.1_html_report.zip"
	      compare="sim_size" delta="1500" />
      <output name="output_summits_bed_file" file="test_MACS2.1.1_summits.bed" />
      <output name="output_narrowpeaks_file" file="test_MACS2.1.1_peaks_narrowPeak.interval" />
      <output name="output_xls_to_interval_peaks_file"
	      file="test_MACS2.1.1_peaks.xls.re_match"
	      compare="re_match" lines_diff="1" />
      <output name="output_treat_pileup_file" file="test_MACS2.1.1_treat_pileup.bdg" />
      <output name="output_lambda_bedgraph_file" file="test_MACS2.1.1_control_lambda.bdg" />
    </test>
    <!-- Peak calling with bigwig output -->
    <test>
      <!-- Inputs -->
      <param name="experiment_name" value="test_MACS2.1.1" />
      <param name="broad_regions" value="" />
      <param name="format" value="BED" />
      <param name="input_chipseq_file1" value="test_region_IP.bed" dbkey="galGal3"
	     ftype="bed" />
      <param name="input_control_file1" value="test_region_Input.bed"
	     ftype="bed" />
      <param name="gsize" value="" />
      <param name="user_defined_gsize" value="775000000.0" />
      <param name="bw" value="300" />
      <param name="xls_to_interval" value="true" />
      <param name="bdg_options|bdg" value="-B" />
      <param name="bdg_options|spmr" value="--SPMR" />
      <param name="bdg_options|make_bigwig" value="true" />
      <param name="pq_options_selector" value="qvalue" />
      <param name="qvalue" value="0.05" />
      <param name="advanced_options_selector" value="true" />
      <param name="advanced_options|mfoldlo" value="5" />
      <param name="advanced_options|mfoldhi" value="50" />
      <param name="advanced_options|nolambda" value="" />
      <param name="advanced_options|call_summits" value="" />
      <param name="advanced_options|keep_duplicates" value="" />
      <param name="advanced_options|maximum_tags" value="1" />
      <param name="nomodel_type_selector" value="nomodel" />
      <param name="nomodel_type|extsize" value="243" />
      <!-- Outputs -->
      <output name="output_extra_files" file="test_MACS2.1.1_bw_html_report.zip"
	      compare="sim_size" delta="2500" />
      <output name="output_summits_bed_file" file="test_MACS2.1.1_summits.bed" />
      <output name="output_narrowpeaks_file" file="test_MACS2.1.1_peaks_narrowPeak.interval" />
      <output name="output_xls_to_interval_peaks_file"
	      file="test_MACS2.1.1_peaks.xls.re_match"
	      compare="re_match" lines_diff="1" />
      <output name="output_treat_pileup_file" file="test_MACS2.1.1_treat_pileup.bdg" />
      <output name="output_lambda_bedgraph_file" file="test_MACS2.1.1_control_lambda.bdg" />
      <output name="output_bigwig_file" file="test_MACS2.1.1_treat_pileup.bw"
	      compare="sim_size" />
    </test>
  </tests>
  <help>
**What it does**

MACS (Model-based Analysis of ChIP-seq) 2.1.1 provides algorithms for identifying
transcript factor binding sites. The program can be used either for ChIP-Seq data alone,
or with control sample data to improve specificity.

View the MACS2 documentation at:
https://github.com/taoliu/MACS/blob/master/README.rst

------

**Usage**

The tool interfaces with the **callpeak** function in MACS, which calls peaks from
alignment results.

------

**Credits**

This Galaxy tool was based on the MACS2 tool hosted in the Galaxy toolshed at

 * http://toolshed.g2.bx.psu.edu/view/modencode-dcc/macs2

(specifically the 16:14f378e35191 revision of the tool) which is credited to Ziru
Zhou. This version is a reimplemented version developed within the Bioinformatics
Core Facility at the University of Manchester, which uses more up-to-date Galaxy
syntax and adds some extra features.

The tool runs Tao Liu's MACS2 software:

 * https://github.com/taoliu/MACS

The reference for MACS is:

 * Zhang Y, Liu T, Meyer CA, Eeckhoute J, Johnson DS, Bernstein BE, Nusbaum C,
   Myers RM, Brown M, Li W, Liu XS. Model-based analysis of ChIP-Seq (MACS).
   Genome Biol. 2008;9(9):R137.

Please kindly acknowledge both this Galaxy tool and the MACS2 package if you
use it.
  </help>
  <citations>
    <!--
    See https://wiki.galaxyproject.org/Admin/Tools/ToolConfigSyntax#A.3Ccitations.3E_tag_set
    Can be either DOI or Bibtex
    Use http://www.bioinformatics.org/texmed/ to convert PubMed to Bibtex
    -->
    <citation type="doi">10.1186/gb-2008-9-9-r137</citation>
  </citations>
</tool>