Allelic Depth from VCF

[chrisbala]

I'm trying to estimate the # of reads representing each allele across multiple samples in RNA-seq data.

I've identified SNPs with samtools mpileup and vcftools call, but allelic depth is not provided in the vcf.

I gather that GATK Variant Annotator (-A DepthPerAlleleBySample) might be able to extract this info from the vcf, but GATK seems unhappy with my vcf format

Your input file has a malformed header: unexpected tag count 5 in line <ID=VDB,Number=1,Type=Float,Description="Variant Distance Bias for filtering splice-site artefacts in RNA-seq data (bigger is better)",Version="3">

Does anyone have other suggests for getting per sample/per allele counts? Or a suggestion to get around this formatting issue?

Thanks!

[SylvainL]

I seem to remember that with bcftools, you can have this information in the bcf file...

[chrisbala]

OOps, I meant "bcftools call" not vcftools

[SNPsaurus]

If you pass the DPR option to bcftools call it gives the allele depth (I also do DP for total read depth):
samtools mpileup -gu -t DP,DPR

The output is 0/1:153,0,125:61:20,41,0,0
0/1 allele call
153,0,125 prob of allele call
61 read depth
20,41,0,0 depth of different alleles

[chrisbala]

Allelic Depth

Excellent! I'll give it a shot!