Skip to content

Latest commit

 

History

History
54 lines (37 loc) · 3.09 KB

README.md

File metadata and controls

54 lines (37 loc) · 3.09 KB
Raw

install with conda CI Coverage Status

vartriage

Just a little tool to perform second opinions on variant calls. It takes as input a triage VCF file and list of evidence VCF files. Output (written to stdout) is an opdated version of the triage VCF, where filtered variants that PASS in any of the evidence VCF files are marked as PASS.

$ vartriage --help
usage: vartriage [-h] --triage_vcf TRIAGE_VCF --evidence_vcfs EVIDENCE_VCFS

optional arguments:
  -h, --help            show this help message and exit
  --triage_vcf TRIAGE_VCF
                        VCF file to be triaged
  --evidence_vcfs EVIDENCE_VCFS
                        Comma-separated list of ID:VCF_PATH evidence VCF files

When a variant in un-filtered, two INFO fields are added in the output VCF file: VTOF contains the original filter reason before triage, and VTSO contains a comma-separated list of IDs of evidence VCF files which support the second opinion.

Note that multi-allelic variants are not supported by vartriage. It is thus recommended to first process input VCF files using bcftools norm -m- foo.vcf.gz, which splits multi-allelic variants into biallelic variants.

Example

A typical use-case could be to use Strelka as second opinion to Mutect2.

vartriage --triage_vcf mutect2.vcf.gz --evidence_vcfs STRELKA_SNV:strelka.snvs.vcf.gz,STRELKA_INDEL:strelka.indels.vcf.gz | bgzip -c > triaged.vcf.gz

For example, suppose that a variant in mutect2.vcf.gz is filtered:

chr17   39725187        .       C       T       .       clustered_events        AS_FilterStatus=SITE;AS_SB_TABLE=196,188|31,49;DP=472;ECNT=5;GERMQ=93;MBQ=35,20;MFRL=268,229;MMQ=60,60;MPOS=35;NALOD=2.30;NLOD=58.69;POPAF=6.00;ROQ=93;TLOD=190.21      GT:AD:AF:DP:F1R2:F2R1:SB        0/0:231,0:5.023e-03:231:115,0:116,0:113,118,0,0        0/1:153,80:0.306:233:74,41:79,39:83,70,31,49

If the same variant is called (but not filtered) in strelka.snvs.vcf.gz:

chr17   39725187        .       C       T       .       PASS    SOMATIC;QSS=333;TQSS=2;NT=ref;QSS_NT=3070;TQSS_NT=1;SGT=CC->CT;DP=466;MQ=60.00;MQ0=0;ReadPosRankSum=1.08;SNVSB=0.00;SomaticEVS=19.98    DP:FDP:SDP:SUBDP:AU:CU:GU:TU    232:0:0:0:0,0:232,232:0,0:0,0   232:0:0:0:0,0:151,153:0,0:81,81

then the conclusion after triage is:

chr17   39725187        .       C       T       .       PASS    AS_FilterStatus=SITE;AS_SB_TABLE=196,188|31,49;DP=472;ECNT=5;GERMQ=93;MBQ=35,20;MFRL=268,229;MMQ=60,60;MPOS=35;NALOD=2.3;NLOD=58.69;POPAF=6;ROQ=93;TLOD=190.21;VTSO=STRELKA_SNV;VTOF=clustered_events   GT:AD:AF:DP:F1R2:F2R1:SB       0/0:231,0:0.005023:231:115,0:116,0:113,118,0,0  0/1:153,80:0.306:233:74,41:79,39:83,70,31,49

Installation

The recommended way to install vartriage is by using conda:

$ conda install -c micknudsen vartriage