hard filtering one WGS raw.g.vcf


Im trying to apply hard filter to one WGS.g.vcf (coming form HaplotypeCaller).
The command im using is:
java -jar ${GATK}/GenomeAnalysisTK.jar -R ${reference}/mm10.fa -T SelectVariants -V ${variant}/mouseWGS.g.vcf -selectType SNP -o ${output}/my_mouse.vcf

It produce a "my_mouse.vcf" file with only the vcf header (no SNP entries).
When checking the log file there are non warnings and at the end it looks like:

INFO 16:50:56,991 ProgressMeter - chrY:53156701 2.690567022E9 3.3 h 4.0 s 98.5% 3.3 h 2.9 m
INFO 16:51:26,992 ProgressMeter - chrY_JH584301_random:230401 2.729494067E9 3.3 h 4.0 s 100.0% 3.3 h 4.0 s
INFO 16:51:28,959 SelectVariants - 689216856 records processed.
INFO 16:51:28,997 ProgressMeter - done 2.730871774E9 3.3 h 4.0 s 100.0% 3.3 h 0.0 s

INFO 16:51:28,998 ProgressMeter - Total runtime 11762.70 secs, 196.04 min, 3.27 hours

Done. There were no warn messages.

the GATK version im using is: GATK v3.8-0-ge9d806836
I have checked the memory and it is enough; morover, I have tried different options (--selectTypeToInclude SNP, or --selectTypeToExclude INDEL) but getting the same thing, no warnings and no SNPs....Would you please help me in understanding what is wrong?

Thanks a lot in advance for your support,

Best Answer


  • Hi Sheila,

    SelectVariants worked on the vcf generated with GenotypeGVCF...i thought that it wouldn't be necessary when working with only one whole genome sequence.
    Thanks a lot for your help,


  • SheilaSheila Broad InstituteMember, Broadie, Moderator

    Hi Laura,

    The GVCFs are not supposed to be used in final analysis. They are meant to be an intermediate file in the GVCF workflow. Glad to hear your issue is solved :smile:


Sign In or Register to comment.