annotate varscan_mpileup2snp_from_bam.xml @ 0:0c5cc5763091 draft

planemo upload for repository https://bitbucket.org/EMCbioinf/galaxy-tool-shed-tools commit 0bc9864516071632199ddf9a4ff403893060c99f

<?xml version="1.0" encoding="UTF-8"?>
<tool id="varscan_mpileup2snp_from_bam" name="VarScan2 Call SNPs from BAM" version="2.3.6.a">
    <description>VarScan2 SNP/SNV detection; directly reading *.bam file(s) &amp; using parallel mpileup generation, to avoid unnecessairy I/O overhead and increase performance.</description>
    
    <requirements>
        <requirement type="package" version="0.1.19-a">samtools_parallel_mpileup</requirement>
        <requirement type="package" version="0.1.19">samtools</requirement>
        <requirement type="package" version="2.3.6">varscan</requirement>
    </requirements>
    
    <version_command>java -jar $JAVA_JAR_PATH/VarScan.v2.3.6.jar 2>&amp;1 | head -n 1</version_command>
    
    <command>
        #if $reference_genome_source.source_select == "attribute" and len({ alignment.metadata.dbkey:True for alignment in $alignments }.keys()) != 1
            echo "Invalid number of dbkeys are found: ${ len({ alignment.metadata.dbkey:True for alignment in $alignments }.keys()) }, while only one should be used. Make sure that the alignments are done on the same reference genome and that 'tool-data/all_fasta.loc' is configured properly!" >&amp;2
        #else
            #import os.path
            #for $alignment in $alignments
                <!-- @todo use the existence of $alignment.metadata.bam_index or $alignment.metadata['bam_index'] -->
                #if not os.path.isfile(str($alignment)+".bai")
                 echo "- Indexing alignment file: $alignment.name " ; 
                 samtools index $alignment 2>&amp;1 ; 
                #else
                 echo "- Skiping indexing: $alignment.name " ; 
                #end if
            #end for
            
            #if $mpileup_parallelization.mpileup_parallelization_select == "true"
                samtools-parallel-mpileup mpileup
                -t $mpileup_parallelization.samtools_threads
            #else
                samtools mpileup
            #end if
                -f 
                    #if $reference_genome_source.source_select == "indexed_filtered"
                        "$reference_genome_source.reference_genome"
                    #else if $reference_genome_source.source_select == "indexed_all"
                        "$reference_genome_source.reference_genome"
                    #else if $reference_genome_source.source_select == "history"
                        "$reference_genome_source.reference_genome"
                    #else
                        <!--
                            This is a workaround to obtain the "genome.fa" file that
                            corresponds to the dbkey of the alignments.
                            Because this file is "calculated" during run-time, it can
                            be used in a workflow.
                        -->
                        "${ filter( lambda x: str( x[0] ) == str( { alignment.metadata.dbkey:True for alignment in $alignments }.keys()[0] ), $__app__.tool_data_tables[ 'all_fasta' ].get_fields() )[0][-1] }"
                    #end if
            
            #if $extended_parameters_regions.samtools_regions == "region"
                 -r $extended_parameters_regions.samtools_r
            #elif $extended_parameters_regions.samtools_regions == "regions_file_pos" or $extended_parameters_regions.samtools_regions == "regions_file_bed"
                 -l $extended_parameters_regions.samtools_l
            #end if
            
            #if $extended_parameters.parameters == "extended"
                $extended_parameters.samtools_6
                $extended_parameters.samtools_A
                $extended_parameters.samtools_B
                 -C $extended_parameters.samtools_C
                 -d $extended_parameters.samtools_d
                $extended_parameters.samtools_E
                 -M $extended_parameters.samtools_M
                $extended_parameters.samtools_R
                 -q $extended_parameters.samtools_q
                 -Q $extended_parameters.samtools_Q
                
                 -e $extended_parameters.samtools_e
                 -F $extended_parameters.samtools_F
                 -h $extended_parameters.samtools_h
                $extended_parameters.samtools_I
                 -L $extended_parameters.samtools_L
                 -m $extended_parameters.samtools_m
                 -o $extended_parameters.samtools_o
                $extended_parameters.samtools_p
                 -P $extended_parameters.samtools_P
            #end if
            
            #for $alignment in $alignments
                 ${alignment}
            #end for
             2>stderr_1.txt
            
            #if $mpileup_parallelization.mpileup_parallelization_select == "true"
                #if $mpileup_parallelization.sort_mpileup
                 | sort -k1,1V -k2,2g 
                #end if
            #end if
            
            ## Make for every MPILEUP file an 
            ## http://en.wikipedia.org/wiki/Named_pipe
            
             | java
                     -Xmx64G
                     -jar \$JAVA_JAR_PATH/VarScan.v2.3.6.jar
                         mpileup2snp
             
            #if $extended_parameters.parameters == "extended"
                     --min-coverage     $extended_parameters.varscan_min_coverage
                     --min-reads2       $extended_parameters.varscan_min_reads2
                     --min-avg-qual     $extended_parameters.varscan_min_avg_qual
                     --min-var-freq     $extended_parameters.varscan_min_var_freq
                     --min-freq-for-hom $extended_parameters.varscan_min_freq_for_hom
                     --p-value          $extended_parameters.varscan_p_value
                                        $extended_parameters.varscan_strand_filter 
                                        $extended_parameters.varscan_variants 
            #end if
            
            #if $varscan_output == "vcf" or $varscan_output.value == "vcf"
             --output-vcf 1 
            #end if
            
             2>stderr_2.txt 
             > $snv_output ;
             
             
             echo "---------------[ mpileup generation ]---------------" ;
             cat stderr_1.txt ;
             echo "" ;
             echo "---------------[ VarScan SNP detect ]---------------" ;
             cat stderr_2.txt ;
             echo "" ;
             echo "----------------------------------------------------" ;
        #end if
    </command>
    
    <inputs>
        <param format="bam,sam" multiple="true" name="alignments" type="data" label="Alignment file(s)" help="Mapped reads in BAM or SAM format."/>
        
        <!-- Find out how to access the reference genome from the BAM file(s) -->
        <conditional name="reference_genome_source">
            <param name="source_select" type="select" label="Fasta Source">
                <option value="indexed_filtered">Use a built-in index (which fits your reference)</option>
                <option value="history">Use reference from the history</option>
                <option value="indexed_all">Use a built-in index (entire list) - avoid this option if possible; only useful if you design a workflow</option>
                <option value="attribute">Use a built-in index based on the 'metadata.dbkey' attribute; ideal in workflows</option>
            </param>
            <when value="history">
                <param name="reference_genome" format="fasta" type="data" label="Reference Genome used during alignment (FASTA)" help="Reference genome (genome.fa) that corresponds to the *.bam file." />
            </when>
            <when value="indexed_filtered">
                <param name="reference_genome" type="select" label="Reference Genome used during alignment (FASTA)" >
                    <options from_data_table="all_fasta">
                        <column name="name"  index="2"/>
                        <column name="dbkey" index="1"/>
                        <column name="value" index="3"/><!-- Value is the path of the fasta file -->
                        <filter type="data_meta" ref="alignments" multiple="false" key="dbkey" column="1" />
                        <validator type="no_options" message="No indexes are available for the selected input dataset" />
                    </options>
                </param>
            </when>
            <when value="indexed_all">
                <param name="reference_genome" type="select" label="Reference Genome used during alignment (FASTA)" >
                    <options from_data_table="all_fasta">
                        <column name="name"  index="2"/>
                        <column name="dbkey" index="1"/>
                        <column name="value" index="3"/><!-- Value is the path of the fasta file -->
                        <validator type="no_options" message="No indexes are available for the selected input dataset" />
                    </options>
                </param>
            </when>
            <when value="attribute" />
        </conditional>
        
        <conditional name="extended_parameters_regions">
            <param name="samtools_regions" type="select" label="Region specific parameters" help="Let samtools target specific genomic locations.">
                <option value="entire_genome">Entire genome</option>
                <option value="region">Specific region</option>
                <option value="regions_file_pos">Specific positions (file); list of positions</option>
                <option value="regions_file_bed">Specific regions (file); list of regions in BED</option>
            </param>
            <when value="entire_genome" />
            <when value="region">
                <param type="text" name="samtools_r" label="Samtools: region in which pileup is generated" help="e.g. chrX or chr:pos or chr:start-end" />
            </when>
            <when value="regions_file_pos">
                <param type="data" name="samtools_l" format="tabular" label="Samtools: list of positions (chr pos)" />
            </when>
            <when value="regions_file_bed">
                <param type="data" name="samtools_l" format="bed"     label="Samtools: specific regions (BED)" />
            </when>
        </conditional>
        
        <conditional name="mpileup_parallelization">
            <param name="mpileup_parallelization_select" type="select" label="Use parallelization for the mpileup generation (experimental)" help="Especially if larger numbers of bam/sam files are processed, or the file infrastructure is optimized for IO-paralellization, this feature might improve performance.">
                <option value="false" >False - uses classical samtools</option>
                <option value="true">True - uses (experimental) samtools mpileup-parallel</option>
            </param>
            <when value="false" />
            <when value="true">
                <param type="integer" name="samtools_threads" value="2" min="1" label="Samtools: mpileup threads" />
                <param type="boolean" name="sort_mpileup" truevalue="true" falsevalue="false" label="Sort mpileup file (SLOW)" help="Because parallelization may disrupt the outputs order, sorting can be conveniet for e.g. testing. Notice that this function has only use in a limited number of situations but consumes (much) resources. Only use it if it's really neccesairy." />
            </when>
        </conditional>
        
        <conditional name="extended_parameters">
            <param name="parameters" type="select" label="Advanced parameters" help="For more advanced VarScan and samtools settings.">
                <option value="default">Default settings</option>
                <option value="extended">Extended settings</option>
            </param>
            <when value="default" />
            <when value="extended">
                <param type="boolean" name="samtools_6" falsevalue="" truevalue=" -6" label="Samtools: assume the quality is in the Illumina-1.3+ encoding" />
                <param type="boolean" name="samtools_A" falsevalue="" truevalue=" -A" label="Samtools: count anomalous read pairs" />
                <param type="boolean" name="samtools_B" falsevalue="" truevalue=" -B" label="Samtools: disable BAQ computation" />
                <param type="integer" name="samtools_C" value="0"                     label="Samtools: parameter for adjusting mapQ; 0 to disable [0]" />
                <param type="integer" name="samtools_d" value="250"                   label="Samtools: max per-BAM depth to avoid excessive memory usage [250]" />
                <param type="boolean" name="samtools_E" falsevalue="" truevalue=" -E" label="Samtools: recalculate extended BAQ on the fly thus ignoring existing BQs" />
                <param type="integer" name="samtools_M" value="60"                    label="cap mapping quality at INT [60]" />
                <param type="boolean" name="samtools_R" falsevalue="" truevalue=" -R" label="Samtools: ignore RG tags" />
                <param type="integer" name="samtools_q" value="0"                     label="Samtools: skip alignments with mapQ smaller than INT [0]" />
                <param type="integer" name="samtools_Q" value="13"                    label="Samtools: skip bases with baseQ/BAQ smaller than INT [13]" />
                
                <param type="integer" name="samtools_e" value="20"                    label="Samtools: Phred-scaled gap extension seq error probability [20]" />
                <param type="float"   name="samtools_F" value="0.002"                 label="Samtools: minimum fraction of gapped reads for candidates [0.002]" help="Alias: -F" />
                <param type="integer" name="samtools_h" value="100"                   label="Samtools: coefficient for homopolymer errors [100]" />
                <param type="boolean" name="samtools_I" falsevalue="" truevalue=" -I" label="Samtools: do not perform indel calling" />
                <param type="integer" name="samtools_L" value="250"                   label="Samtools: max per-sample depth for INDEL calling [250]" />
                <param type="integer" name="samtools_m" value="1"                     label="Samtools: minimum gapped reads for indel candidates [1]" help="Alias: -m" />
                <param type="integer" name="samtools_o" value="40"                    label="Samtools: Phred-scaled gap open sequencing error probability [40]" />
                <param type="boolean" name="samtools_p" falsevalue="" truevalue=" -p" label="Samtools: apply -m and -F per-sample to increase sensitivity" />
                <param type="text"    name="samtools_P" value="all"                   label="Samtools: comma separated list of platforms for indels [all]" />
                
                <param type="integer" name="varscan_min_coverage"     value="8"       label="VarScan: Minimum read depth at a position to make a call [8]" />
                <param type="integer" name="varscan_min_reads2"	      value="2"       label="VarScan: PMinimum supporting reads at a position to call variants [2]" />
                <param type="integer" name="varscan_min_avg_qual"     value="15"      label="VarScan: Minimum base quality at a position to count a read [15]" />
                <param type="float"   name="varscan_min_var_freq"     value="0.01"    label="VarScan: minimum fraction of gapped reads for candidates [0.002]" help="Alias: -F" />
                <param type="float"   name="varscan_min_freq_for_hom" value="0.75"    label="VarScan: Minimum frequency to call homozygote [0.75]" />
                <param type="float"   name="varscan_p_value"          value="0.99"    label="VarScan: Default p-value threshold for calling variants [99e-02]" />
                <param type="boolean" name="varscan_strand_filter"    falsevalue=" --strand_filter 0" truevalue=" --strand_filter 1" checked="true"  label="VarScan: Ignore variants with >90% support on one strand [1]" />
                <param type="boolean" name="varscan_variants"         falsevalue=" --variants 0"      truevalue=" --variants 1"      checked="false" label="VarScan: Report only variant (SNP/indel) positions [0]" />
            </when>
        </conditional>
        
        <param name="varscan_output" type="select" label="Output format">
            <option value="vcf">VCF</option>
            <option value="tabular">tabular</option>
        </param>
    </inputs>
    
    <outputs>
        <data format="tabular" name="snv_output" label="${tool.name} on ${', '.join([ str(a.hid)+': '+a.name for a in $alignments ])}">
            <change_format>
                <when input="varscan_output" value="vcf" format="vcf" />
            </change_format>
        </data>
    </outputs>
    
    <tests>
        <test><!-- Use classical samtools -->
            <param name="alignments" value="example.bam" ftype="bam" />
            
            <param name="source_select" value="history" />
            <param name="reference_genome" value="example.fa" ftypet="fasta" />
            
            <param name="samtools_regions" value="entire_genome" />
            
            <param name="mpileup_parallelization_select" value="false" />
            <param name="sort_mpileup" value="true" />
            
            <param name="parameters" value="default" />
            <param name="varscan_output_vcf" value="1" />
            
            
            <output name="snv_output" file="example.vcf" />
        </test>
        <test><!-- Use parallelized samtools - @todo replace with sambamba! -->
            <param name="alignments" value="example.bam" ftype="bam" />
            
            <param name="source_select" value="history" />
            <param name="reference_genome" value="example.fa" ftypet="fasta" />
            
            <param name="samtools_regions" value="entire_genome" />
            
            <param name="mpileup_parallelization_select" value="true" />
            <param name="samtools_threads" value="2" />
            <param name="sort_mpileup" value="true" />
            
            <param name="parameters" value="default" />
            <param name="varscan_output_vcf" value="1" />
            
            
            <output name="snv_output" file="example.vcf" />
        </test>
    </tests>
    
    <help>
**VarScan 2.3.6**

VarScan is a platform-independent mutation caller for targeted, exome, and whole-genome resequencing data generated on Illumina, SOLiD, Life/PGM, Roche/454, and similar instruments. The newest version, VarScan 2, is written in Java, so it runs on most operating systems.
http://dx.doi.org/10.1101/gr.129684.111
http://www.ncbi.nlm.nih.gov/pubmed/19542151

*VarScan* requires mpileup formatted input files, which are generally derived from BAM files. Since mpileup files can become humongous, the interim step of storing it is bypassed. Thus, in this wrapper one or multiple BAM/SAM files go in, get processed into a mpileup file and get directly linked to VarScan.
The samtools package is not able to parallelize the mpileup generation which make it a very slow process.
Other people were aware of this and have written a version that can do parallelization:
https://github.com/mydatascience/parallel-mpileup

Consequently, when a BAM files gets processed by this wrapper, it's processed by *parallel-mpileup* before its send to VarScan.

.. _VarScan: http://varscan.sourceforge.net/

**Input formats**

VarScan2 accepts sequencing alignments in the same, either SAM or BAM format (http://samtools.sourceforge.net/). The alignment files have to be linked to a reference genome by galaxy. This is indicated under every history item with e.g.: *"database: hg19"* for a link to hg19, or *"database: ?"* if the link is missing.

**Installation**

Make sure your reference genomes are properly annotated in "tool-data/all_fasta.loc", and linked to the names of the reference used for alignment.

**License**

* VarScan2.3.6: Non-Profit Open Software License 3.0 (Non-Profit OSL 3.0)
* parallel-mpileup: MIT License (https://github.com/mydatascience/parallel-mpileup/blob/master/samtools-0.1.19/COPYING)

Contact
-------

The tool wrapper has been written by Youri Hoogstrate from the Erasmus
Medical Center (Rotterdam, Netherlands) on behalf of the Translational
Research IT (TraIT) project:

http://www.ctmm.nl/en/programmas/infrastructuren/traitprojecttranslationeleresearch

More tools by the Translational Research IT (TraIT) project can be found
in the following toolsheds:

http://toolshed.dtls.nl/

http://toolshed.g2.bx.psu.edu/

http://testtoolshed.g2.bx.psu.edu/
</help>
    <citations>
        <citation type="doi">10.1101/gr.129684.111</citation>
    </citations>
</tool>