annotate samtools_stats.xml @ 0:0d71d9467847 draft

Uploaded

<tool id="samtools_stats" name="Stats" version="2.0">
    <description>generate statistics for BAM dataset</description>
    <macros>
    <import>macros.xml</import>
  </macros>
    <expand macro="requirements"></expand>
    <expand macro="stdio"></expand>
    <expand macro="version_command"></expand>
    <command><![CDATA[
        #if $use_reference.use_ref_selector == "yes":
            #if $use_reference.reference_source.reference_source_selector == "history":
               ln -s "${use_reference.reference_source.ref_file}" && samtools faidx `basename "${use_reference.reference_source.ref_file}"` && samtools stats
            #else:
                samtools stats
            #end if
        #else:
            samtools stats
        #end if
        "${input_file}"
        --coverage ${coverage_min},${coverage_max},${coverage_step}
        ${remove_dups}
        #if str( $filter_by_flags.filter_flags ) == "filter":
            #if $filter_by_flags.require_flags:
                --required-flag ${sum([int(flag) for flag in str($filter_by_flags.require_flags).split(',')])}
            #end if
            #if $filter_by_flags.exclude_flags:
                --filtering-flag ${sum([int(flag) for flag in str($filter_by_flags.exclude_flags).split(',')])}
            #end if
        #end if
        --GC-depth ${gc_depth}
        --insert-size ${insert_size}

        ## The code below is commented out because using -I/--id options causes the following exception
        ## Samtools-htslib: init_group_id() header parsing not yet implemented

        ##if str($read_group) != "":
        ##    -I "${read_group}"
        ##end if

        #if str($read_length) != "0":
            --read-length "${read_length}"
        #end if
        --most-inserts ${most_inserts}
        --trim-quality ${trim_quality}
        #if $use_reference.use_ref_selector == "yes":
            #if $use_reference.reference_source.reference_source_selector != "history":
                --ref-seq "${use_reference.reference_source.ref_file.fields.path}"
            #else:
                --ref-seq "${use_reference.reference_source.ref_file}"
            #end if
        #end if
        > "${output}"
        #if $split_output.split_output_selector == "yes":
            #set outputs_to_split = str($split_output.generate_tables).split(',')
            && mkdir split && echo ${split_output.generate_tables} &&

            #if 'sn' in $outputs_to_split:
                echo "# Summary Numbers\n"  > "split/Summary numbers.tab" &&
                grep -q ^SN  "${output}" ; if [ $? = 0 ] ; then grep ^SN  "${output}" | cut -f 2- >> "split/Summary numbers.tab"  ; fi &&
            #end if

            #if 'ffq' in $outputs_to_split:
                echo "# Columns correspond to qualities and rows to cycles. First column is the cycle number\n" > "split/First Fragment Qualities.tab" &&
                grep -q ^FFQ "${output}" ; if [ $? = 0 ] ; then grep ^FFQ "${output}" | cut -f 2- >> "split/First Fragment Qualities.tab" ; fi &&
            #end if

            #if 'lfq' in $outputs_to_split:
                echo "# Columns correspond to qualities and rows to cycles. First column is the cycle number" > "split/Last Fragment Qualities.tab" &&
                grep -q ^LFQ "${output}" ; if [ $? = 0 ] ; then grep ^LFQ "${output}" | cut -f 2- >> "split/Last Fragment Qualities.tab" ; fi &&
            #end if

            #if 'mpc' in $outputs_to_split:
                echo "# Columns correspond to qualities, rows to cycles. First column is the cycle number, second is the number of N's and the rest is the number of mismatches" > "split/Mismatches per cycle.tab" &&
                grep -q ^MPC "${output}" ; if [ $? = 0 ] ; then grep ^MPC "${output}" | cut -f 2- >> "split/Mismatches per cycle.tab" ; fi &&
            #end if

            #if 'gcf' in $outputs_to_split:
                echo "# GC Content of first fragments" > "split/GC Content of first fragments.tab" &&
                grep -q ^GCF "${output}" ; if [ $? = 0 ] ; then grep ^GCF "${output}" | cut -f 2- >> "split/GC Content of first fragments.tab" ; fi &&
            #end if

            #if 'gcl' in $outputs_to_split:
                echo "# GC Content of last fragments" > "split/GC Content of last fragments.tab" &&
                grep -q ^GCL "${output}" ; if [ $? = 0 ] ; then grep ^GCL "${output}" | cut -f 2- >> "split/GC Content of last fragments.tab" ; fi &&
            #end if

            #if 'gcc' in $outputs_to_split:
                echo "# ACGT content per cycle. The columns are: cycle, and A,C,G,T counts (percent)" > "split/ACGT content per cycle.tab" &&
                grep -q ^GCC "${output}" ; if [ $? = 0 ] ; then grep ^GCC "${output}" | cut -f 2- >> "split/ACGT content per cycle.tab" ; fi &&
            #end if

            #if 'is' in $outputs_to_split:
                echo "# Insert sizes. The columns are: insert size, pairs total, inward oriented pairs, outward oriented pairs, other pairs" > "split/Insert sizes.tab" &&
                grep -q ^IS  "${output}" ; if [ $? = 0 ] ; then grep ^IS  "${output}" | cut -f 2- >> "split/Insert sizes.tab"  ; fi &&
            #end if

            #if 'rl' in $outputs_to_split:
                echo "# Read lengths. The columns are: read length, count" > "split/Read lengths.tab" &&
                grep -q ^RL  "${output}" ; if [ $? = 0 ] ; then grep ^RL  "${output}" | cut -f 2- >> "split/Read lengths.tab"  ; fi &&
            #end if

            #if 'id' in $outputs_to_split:
                echo "# Indel distribution. The columns are: length, number of insertions, number of deletions" > "split/Indel distribution.tab" &&
                grep -q ^ID  "${output}" ; if [ $? = 0 ] ; then grep ^ID  "${output}" | cut -f 2- >> "split/Indel distribution.tab"  ; fi &&
            #end if

            #if 'ic' in $outputs_to_split:
                echo "# Indels per cycle. The columns are: cycle, number of insertions (fwd), .. (rev) , number of deletions (fwd), .. (rev)" >  "split/Indels per cycle.tab"  &&
                grep -q ^IC  "${output}" ; if [ $? = 0 ] ; then grep ^IC  "${output}" | cut -f 2- >> "split/Indels per cycle.tab"  ; fi &&
            #end if

            #if 'cov' in $outputs_to_split:
                echo "# Coverage distribution" > "split/Coverage distribution.tab" &&
                grep -q ^COV "${output}" ; if [ $? = 0 ] ; then grep ^COV "${output}" | cut -f 2- >> "split/Coverage distribution.tab" ; fi &&
            #end if

            #if 'gcd' in $outputs_to_split:
                echo "# GC-depth. The columns are: GC%, unique sequence percentiles, 10th, 25th, 50th, 75th and 90th depth percentile" > "split/GC depth.tab" &&
                grep -q ^GCD "${output}" ; if [ $? = 0 ] ; then grep ^GCD "${output}" | cut -f 2- >> "split/GC depth.tab" ; fi &&
            #end if

            ## Unix true command below

            true

        #end if
        ]]></command>
    <inputs>
        <param name="input_file" type="data" format="sam,bam" label="BAM file" />
        <param name="coverage_min" type="integer" value="1" label="Minimum coverage" help="minimum coverage value for --coverage option"/>
        <param name="coverage_max" type="integer" value="1000" label="Maximum coverage" help="maximum coverage value for --coverage option"/>
        <param name="coverage_step" type="integer" value="1" label="Coverage step" help="step value for --coverage option"/>
        <param name="remove_dups" type="boolean" truevalue="--remove-dups" falsevalue="" checked="False" label="Exclude reads marked as duplicates" help="--remove-dups; default = False"/>
        <conditional name="split_output">
            <param name="split_output_selector" type="select" label="Output" help="Select between a single output or separate outputs for each statistics">
                <option value="no" selected="True">a single summary file</option>
                <option value="yes">separate datasets for each statistics</option>
            </param>
            <when value="no" />
            <when value="yes">
                <param name="generate_tables" type="select" display="checkboxes" multiple="True" label="Statistics to extract">
                    <option value="sn">Summary numbers</option>
                    <option value="ffq">First Fragment Qualities</option>
                    <option value="lfq">Last Fragment Qualities</option>
                    <option value="mpc">Mismatches per cycle</option>
                    <option value="gcf">GC Content of first fragments</option>
                    <option value="gcl">GC Content of last fragments</option>
                    <option value="gcc">ACGT content per cycle</option>
                    <option value="is">Insert sizes</option>
                    <option value="rl">Read lengths</option>
                    <option value="id">Indel distribution</option>
                    <option value="ic">Indels per cycle</option>
                    <option value="cov">Coverage distribution</option>
                    <option value="gcd">GC depth</option>
                </param>
            </when>
        </conditional>
        <conditional name="filter_by_flags">
            <param name="filter_flags" type="select" label="Set filter by flags" help="-f and -F options">
                <option value="nofilter" selected="True">Do not filter</option>
                <option value="filter">Filter by flags to exclude or require</option>
            </param>
            <when value="filter">
                <param name="require_flags" type="select" display="checkboxes" multiple="True" label="Require" help="-f">
                    <option value="1">Read is paired</option>
                    <option value="2">Read is mapped in a proper pair</option>
                    <option value="4">The read is unmapped</option>
                    <option value="8">The mate is unmapped</option>
                    <option value="16">Read strand</option>
                    <option value="32">Mate strand</option>
                    <option value="64">Read is the first in a pair</option>
                    <option value="128">Read is the second in a pair</option>
                    <option value="256">The alignment or this read is not primary</option>
                    <option value="512">The read fails platform/vendor quality checks</option>
                    <option value="1024">The read is a PCR or optical duplicate</option>
                </param>
                <param name="exclude_flags" type="select" display="checkboxes" multiple="True" label="Exclude" help="-F">
                    <option value="1">Read is paired</option>
                    <option value="2">Read is mapped in a proper pair</option>
                    <option value="4">The read is unmapped</option>
                    <option value="8">The mate is unmapped</option>
                    <option value="16">Read strand</option>
                    <option value="32">Mate strand</option>
                    <option value="64">Read is the first in a pair</option>
                    <option value="128">Read is the second in a pair</option>
                    <option value="256">The alignment or this read is not primary</option>
                    <option value="512">The read fails platform/vendor quality checks</option>
                    <option value="1024">The read is a PCR or optical duplicate</option>
                </param>
            </when>
            <when value="nofilter" />

        </conditional>
        <param name="gc_depth" type="float" value="20000" label="GC-depth bin size" help="--GC-depth; decreasing bin size increases memory requirement; default = 20000.0"/>
        <param name="insert_size" type="integer" value="8000" label="Maximum insert size" help="--insert-size; default = 8000"/>

        <!--
            
            The -I option of samtools stats returns the following message in version 1.2:

            Samtools-htslib: init_group_id() header parsing not yet implemented
            Abort trap: 6
    
            Because of this the section below is commented out until this stats bug is fixed

            <param name="read_group" type="select" optional="true" label="Limit to a specific read group name" > 
                <options>
                    <filter type="data_meta" ref="input_file" key="read_groups" />
                </options>
            </param>

        -->

        <param name="read_length" type="integer" value="0" label="Minimum read length to generate statistics for" help="--read-length; default = no cutoff"/>
        <param name="most_inserts" type="float" value="0.99" label="Report only the main part of inserts" help="--most-inserts; default = 0.99"/>
        <param name="trim_quality" type="integer" value="0" label="BWA trim parameter" help="--trim-quality; default = 0"/>
        
        <conditional name="use_reference">
            <param name="use_ref_selector" type="select" label="Use reference sequence" help="--ref-seq; required for GC-depth and mismatches-per-cycle calculation">
                <option value="yes">Use reference</option>
                <option selected="True" value="no">Do not use reference</option>
            </param>
            <when value="yes">
                <conditional name="reference_source">
                    <param name="reference_source_selector" type="select" label="Choose a reference sequence for GC depth">
                        <option value="cached">Locally cached</option>
                        <option value="history">History</option>
                    </param>
                    <when value="cached">
                        <param name="ref_file" type="select" label="Using genome">
                            <options from_data_table="fasta_indexes" />
                            <filter type="data_meta" ref="input_file" key="dbkey" column="1" />
                        </param>
                    </when>
                    <when value="history">
                        <param name="ref_file" type="data" format="fasta" label="Using file" />
                    </when>
                </conditional>
            </when>
            <when value="no" />
        </conditional>

    </inputs>

    <outputs>
        <data format="tabular" name="output" label="${tool.name} on ${on_string}">
            <discover_datasets pattern="(?P&lt;designation&gt;.+)\.tab" ext="tabular" visible="true" directory="split" />
        </data>
    </outputs>
    <tests>
        <test>
            <param name="input_file" value="samtools_stats_input.bam" ftype="bam" />
            <param name="use_ref_selector" value="yes" />
            <param name="reference_source_selector" value="history" />
            <param name="ref_file" value="samtools_stats_ref.fa" ftype="fasta" />
            <output name="output" file="samtools_stats_out1.tab" ftype="tabular" lines_diff="4" />
        </test>
        <test>
            <param name="input_file" value="samtools_stats_input.bam" ftype="bam" />
            <param name="use_ref_selector" value="yes" />
            <param name="reference_source_selector" value="history" />
            <param name="ref_file" value="samtools_stats_ref.fa" ftype="fasta" />
            <param name="split_output_selector" value="yes" />
            <param name="generate_tables" value="sn,mpc,gcc" />
            <output name="output" file="samtools_stats_out2.tab" lines_diff="4">
                <discovered_dataset designation="Summary numbers" ftype="tabular" file="samtools_stats_out2/sn.tab" />
                <discovered_dataset designation="ACGT content per cycle" ftype="tabular" file="samtools_stats_out2/gcc.tab" />
                <discovered_dataset designation="Mismatches per cycle" ftype="tabular" file="samtools_stats_out2/mpc.tab" />
            </output>
        </test>
    </tests>
    <help><![CDATA[
**What it does**

This tool runs the ``samtools stats`` command. It has the following options::

    -c, --coverage <int>,<int>,<int>    Coverage distribution min,max,step [1,1000,1]
    -d, --remove-dups                   Exclude from statistics reads marked as duplicates
    -f, --required-flag  <str|int>      Required flag, 0 for unset. See also `samtools flags` [0]
    -F, --filtering-flag <str|int>      Filtering flag, 0 for unset. See also `samtools flags` [0]
        --GC-depth <float>              the size of GC-depth bins (decreasing bin size increases memory requirement) [2e4]
    -h, --help                          This help message
    -i, --insert-size <int>             Maximum insert size [8000]
    -I, --id <string>                   Include only listed read group or sample name
    -l, --read-length <int>             Include in the statistics only reads with the given read length []
    -m, --most-inserts <float>          Report only the main part of inserts [0.99]
    -q, --trim-quality <int>            The BWA trimming parameter [0]
    -r, --ref-seq <file>                Reference sequence (required for GC-depth and mismatches-per-cycle calculation). Galaxy
                                        will provide options for selecting a reference cached as this Galaxy instance or choosing
                                        one from history.
   

    ]]></help>
    <expand macro="citations"></expand>
</tool>