We've moved!
This site is now read-only. You can find our new documentation site and support forum for posting questions here.
Be sure to read our welcome blog!

vcf file generated using HaplotypeCaller does not contain dat lines


I typed the following command for finding snps and indels:
java -jar GenomeAnalysisTK.jar -T HaplotypeCaller -R chr21/chr21.fa -I chr21/alignments/human38chr21.sorted.bam -o chr21/humanoutput.raw.snps.indels.vcf

I even got the vcf file but it contains only the header and does not contain the data lines.What maybe wrong?
I hae attached the file containing the stack trace.


  • SheilaSheila Broad InstituteMember, Broadie ✭✭✭✭✭


    It looks like HaplotypeCaller did not find any regions that have variants. I know this from "INFO 10:46:36,131 HaplotypeCaller - Ran local assembly on 0 active regions". Do you know of any sites that should be called variant? Can you post IGV screenshots of those sites?


  • aditya123aditya123 USAMember

    Ya I did the analysis on the same fasta file but using samtools.It shows indels but as I said earlier that it is not showing when gatk is used.I have attached the vcf file generated using samtools for your reference.

  • aditya123aditya123 USAMember

    What maybe the problem?

  • Geraldine_VdAuweraGeraldine_VdAuwera Cambridge, MAMember, Administrator, Broadie admin

    @aditya123 The depth of coverage in your data is extremely low. The HaplotypeCaller is not designed to process such data. Note that the calls you're getting with samtools have very low quality and are not likely to be very reliable. At such low coverage, you have almost zero power to distinguish real variants from sequencing noise. You should probably reevaluate your experimental design.

Sign In or Register to comment.