MuTect2's variants missing QUAL score

Question

[umn_bist]

I was wondering if anyone has had experience with this. I have a set of tumor:matched normal files that need somatic mutation calling done.

I used STAR aligner - and all unique reads were auto-assigned a defined QUAL score (from my understanding, all splice aware aligners do this including TopHat).

Could this be reason why I am not getting a QUAL score? I just have a "." (dot) in the column.

COMMAND:

java -jar -Xmx32g ~/Applications/GenomeAnalysisTK-3.5/GenomeAnalysisTK.jar -T MuTect2 -R /GRCh37/Sequence/WholeGenomeFasta/genome.fa -I:tumor tumor.bam -I:normal normal.bam --dbsnp 00-All%.vcf --cosmic CosmicCodingMuts%.vcf --cosmic CosmicNonCodingVariants%.vcf -L plist_hg19.bed -o raw.vcf --filter_reads_with_N_cigar

OUTPUT:

##reference=file:////GRCh37/Sequence/WholeGenomeFasta/genome.fa
#CHROM  POS ID  REF ALT QUAL    FILTER  INFO    FORMAT  HNSC
1   150936770   .   T   C   .   alt_allele_in_normal    ECNT=1;HCNT=50;MAX_ED=.;MIN_ED=.;NLOD=20.96;TLOD=8.12   GT:AD:AF:ALT_F1R2:ALT_F2R1:FOXOG:QSS:REF_F1R2:REF_F2R1  0/0:120,4:0.033:2:2:0.500:4312,139:64:56
17  7571782 .   T   C   .   alt_allele_in_normal;clustered_events;t_lod_fstar   ECNT=2;HCNT=16;MAX_ED=222;MIN_ED=222;NLOD=29.61;TLOD=5.37   GT:AD:AF:ALT_F1R2:ALT_F2R1:FOXOG:QSS:REF_F1R2:REF_F2R1  0/0:134,3:0.022:2:1:0.667:5246,102:71:63
17  7572004 .   T   C   .   alt_allele_in_normal;clustered_events   ECNT=2;HCNT=21;MAX_ED=222;MIN_ED=222;NLOD=71.37;TLOD=8.48   GT:AD:AF:ALT_F1R2:ALT_F2R1:FOXOG:QSS:REF_F1R2:REF_F2R1  0/0:292,4:0.014:3:1:0.750:11425,160:147:145
17  7572187 .   A   T   .   alt_allele_in_normal;t_lod_fstar    ECNT=1;HCNT=12;MAX_ED=.;MIN_ED=.;NLOD=26.87;TLOD=5.05   GT:AD:AF:ALT_F1R2:ALT_F2R1:FOXOG:QSS:REF_F1R2:REF_F2R1  0/0:124,3:0.024:0:3:1.00:4138,99:59:65
17  7572393 .   A   G   .   alt_allele_in_normal    ECNT=1;HCNT=1;MAX_ED=.;MIN_ED=.;NLOD=79.91;TLOD=8.72    GT:AD:AF:ALT_F1R2:ALT_F2R1:FOXOG:QSS:REF_F1R2:REF_F2R1  0/0:327,5:0.015:3:2:0.400:12545,176:175:152
17  7572544 .   T   C   .   alt_allele_in_normal    ECNT=1;HCNT=2;MAX_ED=.;MIN_ED=.;NLOD=57.97;TLOD=10.35   GT:AD:AF:ALT_F1R2:ALT_F2R1:FOXOG:QSS:REF_F1R2:REF_F2R1  0/0:267,7:0.026:1:6:0.143:10425,230:131:136
17  7572758 .   T   C   .   alt_allele_in_normal    ECNT=1;HCNT=18;MAX_ED=.;MIN_ED=.;NLOD=77.37;TLOD=8.05   GT:AD:AF:ALT_F1R2:ALT_F2R1:FOXOG:QSS:REF_F1R2:REF_F2R1  0/0:317,5:0.016:5:0:1.00:11834,170:159:158
17  7572812 .   T   A   .   alt_allele_in_normal;t_lod_fstar    ECNT=1;HCNT=2;MAX_ED=.;MIN_ED=.;NLOD=37.08;TLOD=4.25    GT:AD:AF:ALT_F1R2:ALT_F2R1:FOXOG:QSS:REF_F1R2:REF_F2R1  0/0:151,2:0.013:1:1:0.500:5407,80:77:74
17  7577092 .   C   T   .   alt_allele_in_normal;t_lod_fstar    ECNT=1;HCNT=6;MAX_ED=.;MIN_ED=.;NLOD=56.00;TLOD=4.38    GT:AD:AF:ALT_F1R2:ALT_F2R1:FOXOG:QSS:REF_F1R2:REF_F2R1  0/0:221,3:0.013:3:0:0.00:8151,99:95:126
17  7579875 .   G   C   .   alt_allele_in_normal    ECNT=1;HCNT=1;MAX_ED=.;MIN_ED=.;NLOD=31.36;TLOD=7.35    GT:AD:AF:ALT_F1R2:ALT_F2R1:FOXOG:QSS:REF_F1R2:REF_F2R1  0/0:156,6:0.037:2:4:0.333:6102,158:76:80

Answer 1

[Len Trigg]

It is probably not computed because the QUAL field is not really so well defined / useful for the task of somatic calling. QUAL is that it is defined to reflect the probability that the site is variant, and although it isn't stated in the VCF spec, the assumption is that this would be in any sample in the VCF. So for example, if the normal sample contained a variant, that should be reflected in the QUAL score. What you are probably more interested in is a specific score as to whether the tumor contains the putative somatic mutation. For MuTect, this would be the t_lod_fstar score, and in other callers it is commonly stored in the SSC field as recommended by the TGCA VCF specification.