Samtools mpileup genotype likelihood no PL field when all individuals are 0/0

Question

[cetin.m]

I am calling SNPS using samtools with the command:

samtools mpileup -g -uf /mnt/NAS/share/Ref/human/hg19_chr/Chr_all.fa -l 1000G_D10KApart2.bed -q 15 -Q 20 -b CEPHbam.list -o CEPH_1000G_10Kapart.bcf

I have 5 individuals in CEPHbam.list I am giving a reference SNP list in the file 1000G_D10KApart2.bed

My problem is as follows:

For the positions that are 0/0 for all 5 individuals, I get no PL field.

For ex: chr1 66715 . A . 966 . DP=98;MQSB=0.00315921;MQ0F=0;AN=10;DP4=78,18,0,0;MQ=28 GT 0/0 0/0 0/0 0/0 0/0

OR chr1 46650 . T . 558 . DP=109;MQSB=0.998858;MQ0F=0;AN=6;DP4=60,46,0,0;MQ=19 GT 0/0 0/0 ./. 0/0 ./.

What I need is this: chr1 691544 . T A 999 . DP=147;VDB=0.00863476;SGB=-18.2072;RPB=0.999674;MQB=0.941943;MQSB=0.997675;BQB=0.999976;MQ0F=0;ICB=0.724138;HOB=0.18;AC=7;AN=10;DP4=19,27,47,42;MQ=35 GT:PL 0/1:236,0,255 1/1:255,49,0 0/1:90,0,199 1/1:255,69,0 0/1:244,0,218

In short, I need the PL field to be shown for all positions.

How can I achieve this?

Answer 1

[Santosh Anand]

0/0 in all samples implies there are only REF alleles and no ALT allele is present. The PL is calculated as the probability of ALT alleles and given as a PHRED-like score (https://gatkforums.broadinstitute.org/gatk/discussion/1268/what-is-a-vcf-and-how-should-i-interpret-it). Since there are no ALT alleles in your data (note that there is a dot in ALT column too), the PL is missing. Usually, you should not keep these lines in the VCF file as they will unnecessary bloat the size of your VCF.