Identifying Heterozygous Variants in VCF file using vcftools or bcftools

Question

[oars]

To identify the number of heterozygous variants in my .vcf file, I used the following linux command in vcftools:

 $ vcftools --vcf SRR1611183.gatk.vcf --het

And returned the following:

INDV    O(HOM)  E(HOM)  N_SITES F
NA12878 0   0.0 1101    0.00000

This is the same as the 'het_RA_count' => 1101 found in the vcf-stats feature; however, vcf-stats also lists 'het_AA_count' => 2, which is not included in the --het command. If I'm seeking the total number of heterozygous variants should I just total them up (to 1103)? Perhaps het_AA is not truly heterozygous?

I'm open to a bcftools solution if its more straightforward.

Answer 1

[Kevin Blighe]

You can try my 'independent' script and see what it returns: A: calculate Per variant Heterozygosity from VCF file

That just counts the occurences of 0/1 1/0 0|1 and 1|0 per line. The total heterozyguos calls in the file will be the total of the final output column.

Another, more complex version of this is here: A: How to get sample names and genotype for SNP in multi-sample VCF file

Note that you should indeed be using BCFtools and not VCFtools. I believe that this advisory is even given by the VCFtools author (Adam).

Kevin

[oars]

Kevin - this is brilliant! Works perfectly, all I needed to do was sum up the 1's in excel.

[Kevin Blighe]

Great - happy boating (oars... boats...)