HaplotypeCaller can't input the known variant from pooled samples

Follow

yangangli

• September 17, 2023 04:18

Hello

a)  used version: gatk-4.0.5.1

b)  used command:

gatk HaplotypeCaller

-R ref.fa

-I 192.q30.sorted.bam

-ERC GVCF

-O 192.g.vcf

-ploidy 192

I pooled many individuals of plant into a pool and add a mutant with known mutation site (191 wild type DNA and 1mutant), I used HaplotypeCaller to call SNP, but gvcf file and vcf file are both don't have the known mutation site. And I used IGVtools to check the bam file before call snp with HaplotypeCaller, I can find the known alternative base at the target site. So why the bam file have the known mutation but HaplotypeCaller can't input this. I don't know how to change the parameters (I tried -stand-call-conf  --minimum-mapping-quality -mbq).

By the way, I used bcftools to call snp and I can find the known mutation site in vcf files, so I think maybe the filters of HaplotypeCaller are so strict with snp?

0

• 

  Gökalp Çelik

  • September 18, 2023 07:33

  Hi yangangli

  Are your samples haploid originally? If not then your ploidy value should be even higher than what you used. For regular diploid samples HaplotypeCaller uses ploidy value of 2 as default. 

  Also can you use the latest version of GATK 4.4.0.0 to test your case? Version 4.0.5.1 is quite old and there has been quite a bit of changes posted along the way. 

  I hope this helps. 

   

  1

  Comment actions Permalink
• 

  yangangli

  • November 03, 2023 13:04

  Thank you for your suggestions, they are very useful to me.

  0

  Comment actions Permalink

Please sign in to leave a comment.