genotype in the output file of bcftools

[pengchy]

I am calling SNP with mpileup and bcftools. A problem of the genotype in the output file of bcftools is encountered. Following is one example line in vcf format.

scaffold30370   21103   .       A       G,T     999     .       DP=2200;VDB=0.0001;AF1=1;AC1=216;DP4=0,0,0,2190;MQ=37;FQ=-45.5  GT:PL:DP:SP:GQ  1/1:120,57,0,120,57,120:19:0:93 1/1:117,69,0,117,69,117:23:0:99 1/1:128,78,0,128,78,128:26:0:99 1/1:109,36,0,109,36,109:12:0:72 1/1:118,63,0,118,63,118:21:0:99 1/1:131,181,0,131,181,131:60:0:99 ...

I have omitted the other individuals. I found that although there are three genotype, as indicated by column 5, the genotype in the 10+ columns only has 0 or 1. As vcf version 4.1 described, there should be genotype 2 in the GT field. I have checked all these snp sites across all 108 individuals.

Is this a bug of bcftools?

[Heisman]

I think you're misinterpreting it. What you are seeing is two alternate alleles, both different from the reference allele. Thus, there are only two bases in the genotype, not three, as the reference allele is not observed in those individuals. It is labeled 1/1 as both of the observed alleles are alternates. In the first individual with this score:

1/1:120,57,0,120,57,120:19:0:9

It implies that the most likely genotype is homozygous for the first alternate allele given, with the second most likely genotypes being heterozygous for wildtype or the other alternate allele (along with the first alternate allele).

[plantae]

According to current VCF definition:
The genotype of this sample. For a diploid, the GT field indicates the two alleles carried by the sample, encoded by a 0 for the REF allele, 1 for the first ALT allele, 2 for the second ALT allele, etc. When there's a single ALT allele (the by far more common case), GT will be either:

0/0 - the sample is homozygous reference
0/1 - the sample is heterozygous, carrying 1 copy of each of the REF and ALT alleles
1/1 - the sample is homozygous alternate

We would expect that 1/2 might present in the output.

I guess samtools only consider two groups of genotype:
0 for reference
1 for any ALT alleles
the likelihood for each genotype could only be traced in PL or GL field.