The current GATK version is 3.7-0
Examples: Monday, today, last week, Mar 26, 3/26/04

#### Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

GATK 3.7 is here! Be sure to read the Version Highlights and optionally the full Release Notes.
Register now for the upcoming GATK Best Practices workshop, Feb 20-22 in Leuven, Belgium. Open to all comers! More info and signup at http://bit.ly/2i4mGxz

# Indel Calling

Member Posts: 6

I have Ion Torrent data. I am trying to call a variant that I know to exist (confirmed with Sanger). In the position where there is the known indel, I have a depth of roughly 80-90 (in two different runs) and a of those between 20-23% of the reads have the insertion called. What parameters should I be adjusting to get this indel to call? I don't mind a large number of false positives.

I've tried several iterations that include indel realignment using known indels (1000G_phase1 and ills_and_1000G_gold_standard) and also excluding them. I have also tried iterations of setting these flags in UnifiedGenotyper:-stand_call_conf 30.0 -stand_emit_conf 0.0 --min_base_quality_score 0 -glm BOTH --dbsnp dbsnp_137.b37.vcf -nt -rf BadCigar -minIndelCnt 3 -minIndelFrac 0.15. I have also attempted to use HaplotypeCaller: -stand_call_conf 30.0 -stand_emit_conf 0.0 --dbsnp dbsnp_137.b37.vcf -rf BadCigar

Any suggestions would be great.

Tagged:

Hi there,

Is your indel quite big? If so you may need to use HaplotypeCaller and override the default ActiveRegion size to increase the callable size.

You can also try running in GENOTYPE_GIVEN_ALLELES mode to force a call.

Geraldine Van der Auwera, PhD

• Member Posts: 6

Thank you Geraldine. I will give this a try. The indel is just a dupT.

• Member Posts: 6

Geraldine,

I used these parameters for UnifiedGenotyper and received no variants and no input indel vcfs for the indel realignment step.

-stand_call_conf 30.0 -stand_emit_conf 0.0 --min_base_quality_score 0 -glm BOTH --dbsnp $dbSNPRef -nt$numThreads -rf BadCigar -minIndelCnt 3 -minIndelFrac 0.15 --genotyping_mode GENOTYPE_GIVEN_ALLELES

I'm obviously missing something. It appears that I should have included the --alleles flag. What should I be passing to the --alleles flag? I don't quite understand RodBinding[VariantContext]. Could I pass a bed file for regions which I want to force a call, so that I can then go back and see what kind of quality scores prevented the dupT from being called in the first place?

Thank you.