We've moved!
This site is now read-only. You can find our new documentation site and support forum for posting questions here.
Be sure to read our welcome blog!

SVAltAligner walker

Geraldine_VdAuweraGeraldine_VdAuwera Cambridge, MAMember, Administrator, Broadie admin
edited September 2012 in GenomeSTRiP Documentation

1. Introduction

The SVAltAligner walker traverses a set of BAM files to compute alignments to
the alternate alleles of structural variations. This walker is one component
of the SVAltAlign pipeline.

2. Inputs / Arguments

  • -I <bam-file> : The set of input BAM files containing records to realign.

  • -altReference <fasta-file> : The fasta file for the alternate allele
    reference sequences. The fasta file must be indexed with 'samtools faidx' or
    the equivalent. This file should be the output from

  • -alignMappedReads : If present, then align all reads in the input BAM files,
    not just unmapped reads (default false).

  • -alignUnmappedMates <true/false> : If true (the default), then align
    unmapped mates of mapped reads (i.e. reads that have a reference position but
    have the unmapped flag set). If set to false, then only reads in the
    "unmapped" portion of the BAM file will be aligned.

  • -md <directory> : The metadata directory containing metadata about the
    input data set.

3. Outputs

  • -O <bam-file> : The output from this walker is a BAM file containing new
    alignments for input reads that align to the alternate allele reference
    sequences. If no output file is specified, the output is in SAM format instead
    and is written to standard output.
Post edited by Geraldine_VdAuwera on


  • Hi,
    I want to align the short reads in bam towards a set of alternative contigs by assembly which contains a set of variants.
    First I try SVAltAligner walker but it seems that the class path of this walker is not specified in the relevant pages and thus I don't have the clue how to run it.
    Then I try the pipeline in GenomeStrip package (svtoolkit_1.04.1228.tar.gz)
    /com/extra/Java/7u5/bin/java -cp ${classpath} ${mx} \
    org.broadinstitute.sting.queue.QCommandLine \
    -S $toolsDir/qscript/SVAltAlign.q \
    -S $toolsDir/qscript/SVQScript.q \
    -gatk $toolsDir/lib/gatk/GenomeAnalysisTK.jar \
    -cp $toolsDir/lib/SVToolkit.jar:$toolsDir/lib/gatk/GenomeAnalysisTK.jar \
    -configFile $toolsDir/conf/genstrip_parameters.txt \
    -tempDir $out/tmpdir \
    -md $out/metadata \
    -I $bam \
    -alignUnmappedMates \
    -vcf $vcf \
    -R $ref \
    -O $out/Simu.20.svRealign.bam \
    -run && echo done
    First the Simu.20.realign.alt.fasta will be generated with a title like this
    ">._1 L:20:13823-14022:1-200|R:20:14038-14237:342-541|LENGTH:541"
    ">._1 L:20:5655407-5655623:1-217|R:20:5655607-5655823:342-558|LENGTH:558"

    Then I met this problem.
    net.sf.picard.PicardException: Sequence name appears more than once in reference: ._1

    I think the space between ">._1" and "L:20..." is the reason for the error. Could you please tell me how to run SVAltAligner walker and help me have a look at the Picardexception problem?

  • bhandsakerbhandsaker Member, Broadie ✭✭✭✭

    Many of the tools and utilities in svtoolkit require that every variant has a unique ID field.
    I suspect your variants don't have IDs (i.e. they are all ".").

Sign In or Register to comment.