GATK multi sample variant calling

Question

[cetin.m]

I am trying to call SNPS from two bam files simultaneously. I want it to be written to the same vcf, with one column for each sample.

I run the following command:

./GenomeAnalysisTK.jar -T UnifiedGenotyper -I sampleA.bam -I sampleB.bam -R /mnt/NEOGENE1/share/ref/genomes/hsa/hs37d5.fa -L /mnt/NAS/projects/2018_MCetin_Selection/imputation/1000G_chr22.bed --output_mode EMIT_ALL_SITES --genotyping_mode GENOTYPE_GIVEN_ALLELES --alleles /mnt/NEOGENE1/share/dna/hsa/genotypes/1000G/ALL.chr22.phase3_shapeit2_mvncall_integrated_v5a.20130502.genotypes.vcf -o output.vcf

The program runs without error, but when I look at the output file, it only contains one sample column, named sample1. (It is possibly sampleB.bam in the input, and sampleA calls without problems individually)

What am I doing wrong?

Thank you for reading!

Answer 1

[Pierre Lindenbaum]

UnifiedGenotyper is deprecated, use HaplotypeCaller

it only contains one sample column, named sample1.

it's because you flagged your bams with the same read-group '@RG/SN:' same attribute named 'sample1' https://gatkforums.broadinstitute.org/gatk/discussion/6472/read-groups

one way to change this is to rename your samples using picard AddOrReplaceReadGroups: https://broadinstitute.github.io/picard/command-line-overview.html#AddOrReplaceReadGroups

[cetin.m]

Makes a lot of sense!