Genome Analysis Toolkit

Variant Discovery in High-Throughput Sequencing Data

GATK process banner

Need Help?

Search our documentation

Community Forum

Hi, How can we help?

Developed in the Data Sciences Platform at the Broad Institute, the toolkit offers a wide variety of tools with a primary focus on variant discovery and genotyping. Its powerful processing engine and high-performance computing features make it capable of taking on projects of any size. Learn more

How to set a COMPRESSION_LEVEL of ApplyBQSR

0

6 comments

  • Avatar
    Bhanu Gandham

    Hi Nickier 

     

    Htsjdk has a property that controls compression level.

    You can use this java option to tweak it:
    --java-options -Dsamjdk.compression_level=X
    where X is 0-9, default is 2 since we found that that is the best balance between file size and time taken to read and write the file.

     

    0
    Comment actions Permalink
  • Avatar
    Nickier

    Hi Bhanu Gandham  ,Thanks a lot ~ I will have a try.

    0
    Comment actions Permalink
  • Avatar
    registered_user

    Compression level 2 effectively doubles the file size. I would be interested to know what kind of speed advantage are we talking about here to outweigh this massive waste of disk space. The BQSR processing takes a long time, so you don't want to re-run it every time you need to use your bam files. Therefore the recalibrated files are the ones you want to keep for long time storage -> MASSIVE WASTE OF DISK SPACE. You can effectively store only half as much samples you could with regular bam files. I am looking into how to compress these monster files now that I have made the mistake of using the default compression level of 2 for a long time.

    0
    Comment actions Permalink
  • Avatar
    Pamela Bretscher

    Hi registered_user,

    Did the solutions you posted in your other thread (https://gatk.broadinstitute.org/hc/en-us/community/posts/4407291540507-ApplyBQSR-wastes-too-much-disk-space) effectively answer the question that you had here?

    Kind regards,

    Pamela

    0
    Comment actions Permalink
  • Avatar
    registered_user

    Hi Pamela Bretscher, yes I think so. Thanks.

    0
    Comment actions Permalink
  • Avatar
    Pamela Bretscher

    registered_user glad to hear!

    0
    Comment actions Permalink

Please sign in to leave a comment.

Powered by Zendesk