bcftools icon indicating copy to clipboard operation
bcftools copied to clipboard

Heterozygous SNP sites calling

Open ettorefedele opened this issue 2 years ago • 3 comments

Hello,

I am currently using the mpileup | bcftools call pipeline to call multiple SNPs in diploid organisms. If I understood this correctly, the call command exploits the genotype likelihoods generated by mpileup (based on a summary of mapped reads on the reference genome) to make the actual calling. I was wondering if there is a way to tweak bcftools to assess the zigosity of SNPs based on a customised AD ratio, such as that all SNPs where ad1/ad2 or ad2/ad1 is >0.2 are called heterozygous. Where ad1 = REF depth and ad2 = ALT depth. Here is an example: #CHROM POS ID REF ALT QUAL FILTER INFO FORMAT sample_sorted.bam chr_1 131328013 . A G 950 . 0 GT:PL:AD:GQ 1/1:168,106,0:779,3193:99

In the example above AD = 779 , 3193 so the fraction between 779 (ad1) and 3193 (ad2) is ca 0.244. In this case I would like to tell bcftools to call this SNP site heterozygous simply based on my criterion, instead it calls it homozygous. Is there a way around this? Thank you.

ettorefedele avatar Mar 15 '22 11:03 ettorefedele

Can you please show also the raw mpileup line? The calling is informed also by the QS tag.

pd3 avatar Mar 29 '22 09:03 pd3

Hi, sorry for the delay in replying.

I am using the older version samtools mpileup | bcftools call . Would that generate the QS tag? I cannot see it in my VCF.

ettorefedele avatar Aug 01 '22 16:08 ettorefedele

You'll need a much newer version of bcftools (ideally the latest) and use bcftools mpileup rather than samtools mpileup. By the way, it always helps to show the full command.

pd3 avatar Aug 15 '22 10:08 pd3