Heads up:
We’re moving the GATK website, docs and forum to a new platform. Read the full story and breakdown of key changes on this blog.
If you happen to see a question you know the answer to, please do chime in and help your fellow community members. We encourage our fourm members to be more involved, jump in and help out your fellow researchers with their questions. GATK forum is a community forum and helping each other with using GATK tools and research is the cornerstone of our success as a genomics research community.We appreciate your help!

Test-drive the GATK tools and Best Practices pipelines on Terra

Check out this blog post to learn how you can get started with GATK and try out the pipelines in preconfigured workspaces (with a user-friendly interface!) without having to install anything.

Picard RevertSam and OUTPUT_BY_READGROUP does not allow specify output files

maartenSmaartenS AmsterdamMember

I am using a pipeline with as input bam and cram files. The first step of this pipeline is RevertSam combined with the option OUTPUT_BY_READGROUP and I walked into the fact that it is impossible to specify your output type of your files and is the same as the input file. (See https://github.com/broadinstitute/picard/blob/50873dab8845bcd0676b6de80f5a3a484a211d58/src/main/java/picard/sam/RevertSam.java#L212 and 278-288) . Since developing two pipelines is not quite time efficient (as is working hours) and converting the cram files into bam files is also not efficient (cpu/io wise) I will focus on a bam only pipeline and specify the output type.

The default is a bam as output and I hide the cram file extension by using the following construct:

cat sample.cram |java -jar picard.jar RevertSam TMP_DIR=. I=/dev/stdin OUTPUT_BY_READGROUP=true SORT_ORDER=queryname O=revertsam COMPRESSION_LEVEL=1 REFERENCE_SEQUENCE=ref.fasta

This smells like a fishy hack to me, but it seems to work. Is there a better/faster/stronger/less hackerish way to do this?


Best Answer


  • SheilaSheila Broad InstituteMember, Broadie admin


    Are you using the latest version of Picard? Can you post the exact command you ran? According to the documentation, RevertSam can output any type of file as long as the extension is properly specified.


  • maartenSmaartenS AmsterdamMember

    I'm using picard 2.60:

    The command's I ran are:
    mkdir revertsam
    java -jar picard.jar RevertSam TMP_DIR=. I=sample.cram OUTPUT_BY_READGROUP=true SORT_ORDER=queryname O=revertsam COMPRESSION_LEVEL=1 REFERENCE_SEQUENCE=ref.fasta

    Because the input is cram the output is also cram:

    The behaviour in line with the documentation:

    """Will output a BAM/SAM file per read group. By default, all outputs will be in BAM format. However, a SAM file will be produced instead for any read group mapped in OUTPUT_MAP to a path ending with '.sam'. A CRAM file will be produced for any read group mapped to a path ending with '.cram'. """

    However, the output file specified with "=O" must be a directory and not filename when OUTPUT_BY_READGROUP is used.

  • SheilaSheila Broad InstituteMember, Broadie admin


    I think if you specify O=revertsam.bam, you should get a BAM file.


  • maartenSmaartenS AmsterdamMember

    No, it does not:
    it gives the following output:

    When OUTPUT_BY_READGROUP=true and OUTPUT is provided, it must be a directory: revertsam.bam

    Which is in line with the documentation.

  • maartenSmaartenS AmsterdamMember
    edited September 2016

    @Geraldine_VdAuwera I will put a feature request at the Picard github repo. I will post the link to the feature request when I made the request.

Sign In or Register to comment.