Why the germline mutation site generated from FilterMutectCalls dont have any base alteration in normal sample?

Follow

jackie chan wang

• August 22, 2023 14:08

Why the germline mutation site generated from FilterMutectCalls do not have any base alteration in the normal sample?
I also use IGV to observe the normal sample bam file, but I can not see any alteration in the germline mutation site.

Thanks for any helpful input in advance.

0

• Official comment

  

  Gökalp Çelik

  • September 06, 2023 15:30

  Hi jackie chan wang

  We have a better response from our team. 

  This is counterintuitive but correct.  Because the population allele frequency is so high, the likelihood ratio of 2^15 for being absent in the germline versus being a germline het does not outweigh the prior odds ratio (roughly a million to one) that overwhelmingly favors germline versus somatic.This is a purely Bayesian calculation and has nothing to do with hard filtering using gnomAD.

  We hope this answers your question. 

  Comment actions Permalink
• 

  jackie chan wang

  • August 22, 2023 15:41

  like this.

  the GATK version is v4.4.0.0
  b) Exact command used: 

  Step1:
  gatk Mutect2 \
  -R /data/index/GATKindex/Homo_sapiens_assembly38.fasta \
  -I /data/breastfq/tmp/lym.sorted.markdup.BQSR.bam \
  -I /data/breastfq/tmpsam/small.sorted.markdup.BQSR.bam \
  -normal lym \
  --germline-resource /data/index/GATKindex/MUTECT_index/af-only-gnomad.hg38.vcf.gz \
  -L /data/breastfq/bed/baits.bed \
  -O ./primary_somatic.vcf.gz \
  -bamout ./small_normal.bam
  
  nohup gatk GetPileupSummaries \
  -I /data/breastfq/tmpsam/small.sorted.markdup.BQSR.bam \
  -V /data/index/GATKindex/MUTECT_index/small_exac_common_3.hg38.vcf.gz \
  -L /data/index/GATKindex/MUTECT_index/small_exac_common_3.hg38.vcf.gz \
  -O ./smallsummary.table &
  
  Step2.1:

  nohup gatk GetPileupSummaries \
  -I /data/breastfq/tmpsam/small.sorted.markdup.BQSR.bam \
  -V /data/index/GATKindex/MUTECT_index/small_exac_common_3.hg38.vcf.gz \
  -L /data/index/GATKindex/MUTECT_index/small_exac_common_3.hg38.vcf.gz \
  -O ./smallsummary.table &
  
  Step2.2:
  gatk GetPileupSummaries \
  -I /data/breastfq/tmp/lym.sorted.markdup.BQSR.bam \
  -V /data/index/GATKindex/MUTECT_index/small_exac_common_3.hg38.vcf.gz \
  -L /data/index/GATKindex/MUTECT_index/small_exac_common_3.hg38.vcf.gz \
  -O ./lymsummary.table
  
  Step3:

  gatk CalculateContamination \
  -I /data/breastfq/smallsummary.table \
  -matched /data/breastfq/privcf/lymsummary.table \
  -tumor-segmentation ./segments.table \
  -O ./pair_calculatecontamination.table
  
  Step4:

  gatk FilterMutectCalls \
  -R /data/index/GATKindex/Homo_sapiens_assembly38.fasta \
  -V /data/breastfq/primary_somatic.vcf.gz \
  --contamination-table /data/breastfq/pair_calculatecontamination.table \
  --stats /data/breastfq/primary_somatic.vcf.gz.stats \
  --tumor-segmentation /data/breastfq/segments.table \
  -O ./somatic_oncefiltered.vcf.gz
  
  

  0

  Comment actions Permalink
• 

  Gökalp Çelik

  • August 28, 2023 09:45

  Hi jackie chan wang

  Do you observe the same filtering applied to the same position when you do not use the resource file from gnomad? It may be possible that site filtering could have been there due to the resource file but not the actual normal bam file you have?

  Can you try your analysis once again without the gnomad resource file ?

  I hope this helps. 

  0

  Comment actions Permalink

Please sign in to leave a comment.