Hi I am confused about interpreting mixed VCF record. I am having 2 nucleotides in the ALT field of the VCF which are separated by comma. The GT field shows that it is homozygous. How should I interpret this? I am pasting the row in the file just as example. Thanks in advance.
chrI 509 . G A,T 119 . DP=60;VDB=8.840780e-04;AF1=1;AC1=2;DP4=0,0,54,3;MQ=20;FQ=-178 GT:PL:GQ 1/1:152,151,0,172,87,151:99
The individual shown here is homozygous (1/1) for the first alternative allele and as such has a AA genotype for this position. The alleles for the second alternative allele would be presented with a '2'. If the individual would have a AT genotype it would be 1/2, if the individual would be heterozygous GT the genotype would be 0/2.
However if the position is homozygous then why do we get a mixed VCF record
It was probably part of a larger vcf which was filtered? I don't know how you obtained this file or how it was generated.
Yes the line posted above is just one of the records of my vcf file "S2_vcf". Below is work-flow I used to obtain the file.
bwa mem genome.fsa S2_1.fq S2_2.fq > S2.sam -t 4
samtools view -S S2.sam -b -o S2_pe
samtools sort S2_pe Sorted_S2_pe
samtools mpileup -g -u -f genome.fsa Sorted_S2_pe.bam > S2_var
bcftools view -c -v -g S2_var > S2_vcf
yes I am using the above commands to generate the vcf file but what is the reason to have mixed vcf record for homozygous position. Or is there an issue with the workflow.
I'm not familiar with using samtools mpileup for variant calling (since I use GATK), so I'm not sure.
Log in to answer this question.
rather a question then a tutorial.
Thank you Wouter. Would you please share any written material/link about interpreting genotypes.
Please use
ADD COMMENTto answer to earlier posts, as such this thread remains logically structured and easy to follow.Among similar resources, the samtools vcf specifications will put you on the right track. You could have found this yourself by googling.
https://samtools.github.io/hts-specs/VCFv4.2.pdf
Our links are particularly similar ;-)