Mutect2 guide for multiple lanes?

Follow

vctrymao

• April 14, 2020 15:30

I am using GATK 4.1.2.0. I have been following this guide (https://gatk.broadinstitute.org/hc/en-us/articles/360035889471-How-should-I-pre-process-data-from-multiplexed-sequencing-and-multi-library-designs- ) in order to run Mutect on samples with multiple lanes. Is there anything else I need to do beyond this in order to prepare the samples to run with Mutect2? I tried running Mutect but I noticed that all of the calls have zero reference depth (ie. all alternate depth), which doesn't seem right since I ran this on a slightly older version of Mutect and got something different.

Both my tumor and normal samples are multiplexed. I followed the guide and got these read groups now for my tumor sample (and similarly for the normal):

@RG ID:DS-bkm-085-N_L001_RG SM:DS-bkm-085-N_L001 LB:lib_name PL:illumina

@RG ID:DS-bkm-085-N_L002_RG SM:DS-bkm-085-N_L002 LB:lib_name PL:illumina

@RG ID:DS-bkm-085-N_L003_RG SM:DS-bkm-085-N_L003 LB:lib_name PL:illumina

@RG ID:DS-bkm-085-N_L004_RG SM:DS-bkm-085-N_L004 LB:lib_name PL:illumina

@RG ID:DS-bkm-085-N_L005_RG SM:DS-bkm-085-N_L005 LB:lib_name PL:illumina

@RG ID:DS-bkm-085-N_L007_RG SM:DS-bkm-085-N_L007 LB:lib_name PL:illumina

 

There was no error from Mutect so I assumed it was fine. Am I doing something wrong?

0

• 

  Daniel von Rhein

  • April 15, 2020 10:04

  Simply merge your fastq files into one big fastq-file.

  In Linux, this could look like this

  zcat DS-bkm-085-N_L00{1,2,3,4,5,6,7}_RG.fastq | gzip > DS-bkm-085-N_merged_RG.fastq
  

   

  0

  Comment actions Permalink
• 

  David Benjamin

  • April 17, 2020 05:05

  vctrymao Currently you have different sample names for each lane, which is activating multisample mode in recent releases of Mutect2 (since 4.1 or maybe 4.1.1, I think).  You should give every tumor read group the same SM tag.

  0

  Comment actions Permalink

Please sign in to leave a comment.