annotate star_fusion.xml @ 1:0b44456754e2 draft

planemo upload for repository https://github.com/galaxyproject/tools-iuc/tree/master/tools/star_fusion commit d6a7537117d814677113ba9a8e4783a80dd228a2

<tool id="star_fusion" name="STAR-Fusion" version="0.5.4-3" profile="17.01">
    <description>detect fusion genes in RNA-Seq data</description>
    <requirements>
        <!-- Bio-conda -->
        <requirement type="package" version="0.5.4">star-fusion</requirement>
    </requirements>

    <stdio>
        <regex match="command not found" source="stderr" level="fatal"/>
        <regex match="EXITING because of INPUT ERROR" source="stderr" level="fatal"/>
        <regex match="FATAL ERROR" source="stderr" level="fatal"/>
        
        <regex match="Warning:" source="stderr" level="warning"/>
        <regex match="CMD:" source="stderr" level="warning"/>
        
        <regex match="-done creating index file:" source="stderr" level="warning"/>
        <regex match="-parsing GTF file:" source="stderr" level="warning"/>
        <regex match="-building interval tree" source="stderr" level="warning"/>
        <regex match="-parsing fusion evidence:" source="stderr" level="warning"/>
        <regex match="-mapping reads to genes" source="stderr" level="warning"/>
        <regex match="-outputting fusion candidates to file:" source="stderr" level="warning"/>
        
        <regex match="Process complete" source="stderr" level="warning"/>
    </stdio>

    <version_command>STAR-Fusion --version  2>&amp;1 | grep version | grep -o -E "software version.*?"</version_command>

    <command><![CDATA[
        ## 1. ensure the blastn file is provided as *.gz
        if file --mime-type '${blast_pairs}' | grep -q /gzip\$; then
            gzip_suffix='' ;
        else
            ## Older versions of gzip do not support the -k option to keep
            ## the original file - this should be an universion solution
            
            gzip -1 -c -- '${blast_pairs}' > '${blast_pairs}.gz' &&
            gzip_suffix='.gz' ;
        fi &&
        
        ## 2. create reference index - using \$(pwd) is necessary, probably because the perl script changes work directory
        ## - @todo once write a decent STAR and STAR Fusion data manager
        prep_genome_lib.pl
            --genome_fa '${fasta_type.ownFile}'
            --gtf '${geneModel}'
            --blast_pairs "${blast_pairs}\$gzip_suffix"
            --CPU \${GALAXY_SLOTS:-1}
            --output_dir "\$(pwd)/tmp_star_fusion_genome_dir"
        &&

        ## Link in fastq files so they have appropriate extensions
        #if str($input_params.input_source) != "use_chimeric":
            #if $input_params.left_fq.is_of_type("fastq.gz"):
                #set read1 = 'input_1.fastq.gz'
            #else:
                #set read1 = 'input_1.fastq'
            #end if
            ln -f -s '${input_params.left_fq}' ${read1} &&

            #if $input_params.right_fq:
                #if $input_params.right_fq.is_of_type("fastq.gz"):
                    #set read2 = 'input_2.fastq.gz'
                #else:
                    #set read2 = 'input_2.fastq'
                #end if
                ln -f -s '${input_params.right_fq}' ${read2} &&
            #end if
        #end if
        
        ## 3. Run STAR-Fusion
        STAR-Fusion
            #if str($input_params.input_source) == "use_chimeric":
                --chimeric_junction '${input_params.chimeric_junction}'
            #else:
                --left_fq ${read1}
                #if $input_params.right_fq:
                    --right_fq ${read2}
                #end if
            #end if

            --genome_lib_dir "\$(pwd)/tmp_star_fusion_genome_dir"

        #if str($params.settingsType) == "full":
            --min_junction_reads $params.min_junction_reads
            --min_sum_frags $params.min_sum_frags
            --max_promiscuity $params.max_promiscuity
            --min_novel_junction_support $params.min_novel_junction_support
            --min_alt_pct_junction $params.min_alt_pct_junction
            --aggregate_novel_junction_dist $params.aggregate_novel_junction_dist
            --E $params.E
        #end if
    ]]></command>

    <inputs>
        <conditional name="input_params">
            <param name="input_source"
                   type="select"
                   label="Use output from earlier STAR run or let STAR Fusion control running STAR">
                <option value="use_chimeric">Use output from earlier STAR</option>
                <option value="use_fastq">Let STAR Fusion control running STAR</option>
            </param>
            <when value="use_chimeric">
                <param name="chimeric_junction"
                       type="data"
                       format="interval"
                       argument="--chimeric_junction"
                       label="Chimeric junction file from STAR (with STAR-Fusion settings)"/>
            </when>
            <when value="use_fastq">
                <param name="left_fq"
                       type="data"
                       format="fastqsanger,fastqsanger.gz"
                       argument="--left_fq"
                       label="left.fq file"/>
                <param name="right_fq"
                       type="data"
                       format="fastqsanger,fastqsanger.gz"
                       optional="true"
                       argument="--right_fq"
                       label="right.fq file (actually optional, but highly recommended)"/>
            </when>
        </conditional>

        <!-- Genome source. -->
        <conditional name="fasta_type">
            <param name="fasta_type_selector" type="select" label="Source for sequence to search">
                <option value="cached">Locally Cached sequences</option>
                <option value="history" selected="true">Sequences from your history</option>
            </param>
            <when value="cached">
                <param name="ownFile"
                       type="select" label="Genome to search">
                    <options from_data_table="all_fasta">
                        <column name="dbkey" index="1"/>
                        <column name="name" index="2"/>
                        <column name="value" index="3"/>
                    </options>
                </param>
            </when>
            <when value="history">
                <param name="ownFile"
                       type="data"
                       format="fasta"
                       label="Select the reference genome (FASTA file)"/>
            </when>
        </conditional>
        
        <param name="geneModel"
               type="data"
               format="gff3,gtf"
               label="Gene model (gff3,gtf) file for splice junctions and fusion gene detection"/>
        <param name="blast_pairs"
               type="data"
               format="tabular"
               label="Result of BLAST+-blastn of the reference fasta sequence with itself"
               help="Run blastn with '-outputfmt 6' or choose 'Tabular (standard 12 columns)' in the Galaxy wrapper. For human data it is recommended to obtain it from here: https://data.broadinstitute.org/Trinity/CTAT_RESOURCE_LIB/"/>
        
        <conditional name="params">
            <param name="settingsType" type="select" label="Settings to use" help="You can use the default settings or set custom values for any STAR Fusion parameter.">
                <option value="default" selected="true">Use Defaults</option>
                <option value="full">Full parameter list</option>
            </param>
            <when value="default" />
            <when value="full"><!-- Full/advanced params. -->
                <param name="min_junction_reads"
                       type="integer" value="1"
                       label="minimum number of junction-spanning reads required."
                       argument="--min_junction_reads"/>
                <param name="min_sum_frags"
                       type="integer"
                       value="2"
                       label="minimum fusion support = (#junction_reads + #spanning_frags)"
                       argument="--min_sum_frags"/>
                <param name="max_promiscuity"
                       type="integer"
                       value="3"
                       label="maximum number of partners allowed for a given fusion"
                       argument="--max_promiscuity"/>
                <param name="min_novel_junction_support"
                       type="integer"
                       value="3"
                       label="minimum of 3 junction reads required if breakpoint lacks involvement of only reference junctions"
                       argument="--min_novel_junction_support"/>
                <param name="min_alt_pct_junction"
                       type="float"
                       value="10"
                       label="10% of the dominant isoform junction support"
                       argument="--min_alt_pct_junction"/>
                <param name="aggregate_novel_junction_dist"
                       type="integer"
                       value="5"
                       label="non-ref junctions within 5 are merged into single calls"
                       argument="--aggregate_novel_junction_dist"/>
                <param name="E"
                       type="float"
                       value="0.001"
                       label="E-value threshold for blast searches"
                       argument="-E"/>
            </when>
        </conditional>
    </inputs>

    <outputs>
        <data format="tabular" name="output_final" label="${tool.name} on ${on_string}: fusion_candidates.final" from_work_dir="star-fusion.fusion_candidates.final"/>
    </outputs>

    <tests>
        <test>
            <param name="input_source" value="use_chimeric" />
            <param name="chimeric_junction" ftype="interval" value="test1.tabular" />
            <param name="fasta_type_selector" value="history" />
            <param name="ownFile" ftype="fasta" value="test1.fa" />
            <param name="geneModel" ftype="gtf" value="test1.gtf" />
            <param name="blast_pairs" ftype="tabular" value="test1-test1.blastn.tabular" />
            <param name="settingsType" value="default" />
            
            <!-- Last column of the results contains data in a random order so exact matching is not feasible -->
            <output name="output_final">
                <assert_contents>
                    <has_line line="#fusion_name&#009;JunctionReads&#009;SpanningFrags&#009;Splice_type&#009;LeftGene&#009;LeftBreakpoint&#009;RightGene&#009;RightBreakpoint&#009;JunctionReads&#009;SpanningFrags" />
                    <has_text text="GENE1--GENE2&#009;24&#009;0&#009;INCL_NON_REF_SPLICE&#009;GENE1^GENE1&#009;chr1:240:+&#009;GENE2^GENE2&#009;chr2:241:+" />
                </assert_contents>
            </output>
        </test>
        <test>
            <param name="input_source" value="use_fastq" />
            <param name="left_fq" ftype="fastqsanger" value="test1.fastqsanger"/>
            <param name="fasta_type_selector" value="history" />
            <param name="ownFile" ftype="fasta" value="test1.fa" />
            <param name="geneModel" ftype="gtf" value="test1.gtf" />
            <param name="blast_pairs" ftype="tabular" value="test1-test1.blastn.tabular" />
            <param name="settingsType" value="default" />
            
            <!-- Last column of the results contains data in a random order so exact matching is not feasible -->
            <output name="output_final">
                <assert_contents>
                    <has_line line="#fusion_name&#009;JunctionReads&#009;SpanningFrags&#009;Splice_type&#009;LeftGene&#009;LeftBreakpoint&#009;RightGene&#009;RightBreakpoint&#009;JunctionReads&#009;SpanningFrags" />
                    <has_text text="GENE1--GENE2&#009;24&#009;0&#009;INCL_NON_REF_SPLICE&#009;GENE1^GENE1&#009;chr1:240:+&#009;GENE2^GENE2&#009;chr2:241:+" />
                </assert_contents>
            </output>
        </test>
        <test>
            <param name="input_source" value="use_fastq" />
            <param name="left_fq" ftype="fastqsanger.gz" value="test1.fastqsanger.gz"/>
            <param name="fasta_type_selector" value="history" />
            <param name="ownFile" ftype="fasta" value="test1.fa" />
            <param name="geneModel" ftype="gtf" value="test1.gtf" />
            <param name="blast_pairs" ftype="tabular" value="test1-test1.blastn.tabular" />
            <param name="settingsType" value="default" />
            
            <!-- Last column of the results contains data in a random order so exact matching is not feasible -->
            <output name="output_final">
                <assert_contents>
                    <has_line line="#fusion_name&#009;JunctionReads&#009;SpanningFrags&#009;Splice_type&#009;LeftGene&#009;LeftBreakpoint&#009;RightGene&#009;RightBreakpoint&#009;JunctionReads&#009;SpanningFrags" />
                    <has_text text="GENE1--GENE2&#009;24&#009;0&#009;INCL_NON_REF_SPLICE&#009;GENE1^GENE1&#009;chr1:240:+&#009;GENE2^GENE2&#009;chr2:241:+" />
                </assert_contents>
            </output>
        </test>
    </tests>
    <help>
**What it does**

STAR-Fusion is a component of the Trinity Cancer Transcriptome Analysis Toolkit (CTAT). STAR-Fusion uses the STAR aligner to identify candidate fusion transcripts supported by Illumina reads. STAR-Fusion further processes the output generated by the STAR aligner to map junction reads and spanning reads to a reference annotation set.

**Input: files required to run STAR-Fusion**
 - A genome reference sequence (FASTA-format)
 - A corresponding protein-coding gene annotation set (GTF/GFF Format)
 - A last-matching gene pairs file - in Galaxy you can create such files with the *ncbi_blast_plus* tool suite containing *blastn*: https://toolshed.g2.bx.psu.edu/view/devteam/ncbi_blast_plus
 - A STAR chimeric/junction output file - this is optional as STAR Fusion can control running STAR as well.

The authors of STAR Fusion have made some of these files avaialble at: https://data.broadinstitute.org/Trinity/CTAT_RESOURCE_LIB/. The gene annotations in each case are restricted to the protein-coding and lincRNA transcripts.
More info: https://github.com/STAR-Fusion/STAR-Fusion/wiki

    </help>

    <citations>
        <citation type="bibtex">
           @unpublished{star_fusion,
              author = {Brian Haas and Nicolas Stransky and Daniel Nicorici}, 
              title  = {STAR-Fusion},
              url    = {https://github.com/STAR-Fusion/STAR-Fusion}
            }
        </citation>
    </citations>
</tool>