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!

defining read groups for multiple lanes

aushaush Member
edited April 2019 in Ask the GATK team
In the fastq files that I get from NGS provider, each fastq file - corresponding to one sample - contains reads from two lanes of the same flowcell, i.e. I have two types of read names:
In the "Read groups" document (gatkforums.broadinstitute.org/gatk/discussion/6472/read-groups/), it is suggested to create a separate bam file for each lane. Is it the only way to define read groups correctly? And if yes, I'll have to first split the original fastq file according to the lanes?

P.S. a bit offtopic: there is nice section "Deriving ID and PU fields from read names" in that document - but it's only for the old format of the read names. Now Illumina machines use different format: @instrument:Run:FlowcellID:Lane:Tile:X:Y. It would be very helpful if someone could update the document with the similar example using the current format...


  • aushaush Member
    oh, and even more important question on the same subject: when I have multiple bam files (multiple read groups due to multiple lanes) for the same sample - how I proceed it for the Calling variants (software.broadinstitute.org/gatk/documentation/article.php?id=3891)? On which step I can combine it?
  • bhanuGandhambhanuGandham Cambridge MAMember, Administrator, Broadie, Moderator admin

    HI @aush

    You don't necessarily need different physical bams for different read groups. As long as your bam file has read group information, that is sufficient.

Sign In or Register to comment.