Variant not passing VarDict filters

aldosr commented 2 years ago

Hello, I'm having issue with a particular variant, which isn't called by VarDict using the following command:

vardict \
        -G ${FASTA} \
        -N ${NAME} \
        -f ${min_af} \
        -b ${INPUTBAM} \
        -z -c 1 -S 2 -E 3 -g 4 -th 4 \
        ${BED} \
        | teststrandbias.R \
        | var2vcf_valid.pl -N ${NAME} -E -f ${min_af} \
        | awk '{if ($1 ~ /^#/) print; else if ($4 !=5) print}' \
        > ${OUTDIR}/${NAME}.vcf

with min_af=0.0001

However, using the -p command, the variant is recognized as:

chr17 7674291 . C A 77 f0.001;SN1.5 SAMPLE=21752_PAP_T0_S31;TYPE=SNV;DP=17669;VD=9;AF=0.0005;BIAS=2:2;REFBIAS=9613:8039;VARBIAS=1:8;PMEAN=37.6;PSTD=1;QUAL=24.6;QSTD=1;SBF=0;ODDRATIO=143.75054;MQ=60.0;SN=0.800;HIAF=0.0002;ADJAF=0;SHIFT3=0;MSI=1.000;MSILEN=1;NM=3.1;HICNT=4;HICOV=16701;LSEQ=GTGGTACAGTCAGAGCCAAC;RSEQ=TAGGAGATAACACAGGCCCA;DUPRATE=0.625;SPLITREAD=0;SPANPAIR=0GT:DP:VD:AD:AF:RD:ALD 0/0:17669:9:17652,9:0.0005:9613,8039:1,8

I don't get which filter this variant is not passing, let me know if you need the related BAM. Thanks

PolinaBevad commented 2 years ago

Hi @aldosr, Sorry for the late reply!

It seems that this variant has low SN (ratio of high-quality reads to low-quality reads) - here It is 0.8. Threshold is 1.5. Can you try to adjust it to see if the variant passes the filter?

Thanks.

aldosr commented 2 years ago

Hi @PolinaBevad, Thanks for the answer and sorry for the late reply. I didn't check that variant because, after following analyses, I was not sure that variant was "real", and I ended up not needing it for the initial purpose of the experiment. Anyway, I'm gonna check as soon as I can if changing the SN ratio will make the call (can be interesting for other cases). Thanks again, I close the issue, will reopen if needed.

AstraZeneca-NGS / VarDictJava

Variant not passing VarDict filters #349