Mutect2 variant calling output with tumor and matched normal: 0/1 in all variants called
I'm using Mutect2 (GATK 4.1.4.1) to call somatic mutations in tumor and matched normal samples. However, for all the positions called by the tool, a 0/1 genotype is reported for both tumor and normal samples. I've also tried to call variants where a read depth of at least 30 is observed, although those positions having a read depth below 30 are still being called.
I'm using this command:
./gatk-4.1.4.1/gatk Mutect2 -R ./human_g1k_v37.fasta -I Tumor.bam -I Normal.bam --callable-depth 30 -O Sample_somatic.vcf.gz
And the first lines of the output:
#CHROM POS ID REF ALT QUAL FILTER INFO FORMAT ESO08N ESO08T
chr1 1178536 . C CCATCACCGAACTT . . DP=6;ECNT=1;MBQ=0,33;MFRL=0,0;MMQ=60,90;MPOS=0;POPAF=6.00;TLOD=20.95 GT:AD:AF:DP:F1R2:F2R1:SB 0/1:0,6:0.875:6:0,4:0,2:0,0,2,4 0/1:0,0:0.500:0:0,0:0,0:0,0,0,0
chr1 3517109 . G A . . DP=2;ECNT=2;MBQ=27,35;MFRL=0,0;MMQ=96,90;MPOS=41;POPAF=6.00;TLOD=3.72 GT:AD:AF:DP:F1R2:F2R1:PGT:PID:PS:SB 0|1:0,1:0.667:1:0,0:0,1:0|1:3517109_G_A:3517109:0,0,1,0 0|1:1,0:0.333:1:0,0:1,0:0|1:3517109_G_A:3517109:1,0,0,0
chr1 3517115 . G A . . DP=2;ECNT=2;MBQ=33,29;MFRL=0,0;MMQ=96,90;MPOS=47;POPAF=6.00;TLOD=3.72 GT:AD:AF:DP:F1R2:F2R1:PGT:PID:PS:SB 0|1:0,1:0.667:1:0,0:0,1:0|1:3517109_G_A:3517109:0,0,1,0 0|1:1,0:0.333:1:0,0:1,0:0|1:3517109_G_A:3517109:1,0,0,0
chr1 4613604 . C T . . DP=17;ECNT=1;MBQ=32,33;MFRL=0,0;MMQ=94,91;MPOS=45;POPAF=6.00;TLOD=7.24 GT:AD:AF:DP:F1R2:F2R1:SB 0/1:7,2:0.273:9:5,2:1,0:1,6,0,2 0/1:7,1:0.200:8:4,1:3,0:3,4,0,1
chr1 5553227 . C A . . DP=1;ECNT=1;MBQ=0,31;MFRL=0,0;MMQ=60,78;MPOS=9;POPAF=6.00;TLOD=3.28 GT:AD:AF:DP:F1R2:F2R1:SB 0/1:0,1:0.667:1:0,1:0,0:0,0,0,1 0/1:0,0:0.500:0:0,0:0,0:0,0,0,0
chr1 6120721 . G A . . DP=1;ECNT=2;MBQ=0,33;MFRL=0,0;MMQ=60,50;MPOS=62;POPAF=6.00;TLOD=3.48 GT:AD:AF:DP:F1R2:F2R1:SB 0/1:0,0:0.500:0:0,0:0,0:0,0,0,0 0/1:0,1:0.667:1:0,1:0,0:0,0,0,1
chr1 6120734 . G A . . DP=1;ECNT=2;MBQ=0,34;MFRL=0,0;MMQ=60,50;MPOS=75;POPAF=6.00;TLOD=3.58 GT:AD:AF:DP:F1R2:F2R1:SB 0/1:0,0:0.500:0:0,0:0,0:0,0,0,0 0/1:0,1:0.667:1:0,1:0,0:0,0,0,1
-
@RicPin You need to specify which sample is normal, in this case with `-normal ESO08N`. Mutect2 treats any sample not specified as normal as a tumor. That is, you were unwittingly running Mutect2 in multiple-tumor mode.
By the way, `--callable-depth` does something different from what you would expect. The Javadoc describes it as: "Minimum depth to be considered callable for Mutect stats. Does not affect genotyping."
Also, we strongly recommend using our best practices resources for the `--germline-resource` and `--pon` (panel of normals) arguments. You can obtain them at gs://gatk-best-practices/somatic-b37/Mutect2-WGS-panel-b37.vcf / gs://gatk-best-practices/somatic-b37/Mutect2-exome-panel.vcf and gs://gatk-best-practices/somatic-b37/af-only-gnomad.raw.sites.vcf.
Finally, don't forget to run FilterMutectCalls. Mutect2 outputs unfiltered calls, the great majority of which are false positives.
-
Thanks for your reply @David Benjamin. It has helped a lot.
However, the links you've posted before are suitable for the b37 reference genome, and I am working with the h19. Can I use them anyway in my analyses?
Best regards
-
The resources will work with hg19. Occasionally one encounters versions with a "chr1" vs "1" naming convention. In such cases, you can manually delete all "chrs" -- for example, in vim the command :%s/chr//g does the job -- and then reindex.
-
Sorry, but I can't find the file gs://gatk-best-practices/somatic-b37/Mutect2-exome-panel.vcf in GATK FTP Server. Is that the correct filename?
-
Anything starting with gs:// is in a google cloud bucket. GATK tools can read from them directly, or you can download them with gsutil cp.
Please sign in to leave a comment.
5 comments