In a single sample vcf I have homozygous snp that is not assigned a genotype, the genotype is missing (gt=./.).
What knobs can I turn on HaplotypeCaller (in GATK 3.8 or 4.1.2) to assign a genotype to this SNP? What is the threshold that determines whether a genotype is emitted for a variant? I've played around with some flags but haven't hit on the right one yet.
I have a separate sample (different individual) that had a heterozygous SNP at this position and the correct genotype was assigned, the read depth was similar in both cases, the IGV screenshot of both bams is included with this post and this is my HaplotypeCaller command:
```
java -jar -Xmx4g ${GATK} -T HaplotypeCaller \
-R ${ref_fa} \
-I ${PRE}${CHR}.bam \
-L ${bedPadded} \
--genotyping_mode DISCOVERY \
-stand_call_conf 30 \
-o ${PRE}${CHR}.vcf
```
These are the lines from the VCFs (the weird AD:DP:GQ:PL values of 1,1:1:1:1,1,1were added after haplotype caller for downstream annotation software compatibility, in the initial VCF those values are just missing and there's only GT=./.)
Homozygous call
chr2 48018081 . A G 1081.78 PASS AC=0;AF=0;AN=0;BaseQRankSum=-2.065;ClippingRankSum=0;DP=82;ExcessHet=3.0103;FS=0;MLEAC=1;MLEAF=0.5;MQ=60;MQRankSum=0;QD=26.38;ReadPosRankSum=-1.277;SOR=0.693;set=Intersection;type=snps;BUDDY=0 GT:AD:DP:GQ:PL ./.:1,1:1:1:1,1,1
Het call
chr2 48018081 . A G 851.77 PASS AC=1;AF=0.5;AN=2;ClippingRankSum=0;DP=123;ExcessHet=3.0103;FS=0;MLEAC=1;MLEAF=0.5;MQ=60;MQRankSum=0;set=Intersection;type=snps;BUDDY=0 GT:AD:DP:GQ:PL 0/1:29,33:62:99:880,0,869
What knobs can I turn on HaplotypeCaller (in GATK 3.8 or 4.1.2) to assign a genotype to this SNP? What is the threshold that determines whether a genotype is emitted for a variant? I've played around with some flags but haven't hit on the right one yet.
I have a separate sample (different individual) that had a heterozygous SNP at this position and the correct genotype was assigned, the read depth was similar in both cases, the IGV screenshot of both bams is included with this post and this is my HaplotypeCaller command:
```
java -jar -Xmx4g ${GATK} -T HaplotypeCaller \
-R ${ref_fa} \
-I ${PRE}${CHR}.bam \
-L ${bedPadded} \
--genotyping_mode DISCOVERY \
-stand_call_conf 30 \
-o ${PRE}${CHR}.vcf
```
These are the lines from the VCFs (the weird AD:DP:GQ:PL values of 1,1:1:1:1,1,1were added after haplotype caller for downstream annotation software compatibility, in the initial VCF those values are just missing and there's only GT=./.)
Homozygous call
chr2 48018081 . A G 1081.78 PASS AC=0;AF=0;AN=0;BaseQRankSum=-2.065;ClippingRankSum=0;DP=82;ExcessHet=3.0103;FS=0;MLEAC=1;MLEAF=0.5;MQ=60;MQRankSum=0;QD=26.38;ReadPosRankSum=-1.277;SOR=0.693;set=Intersection;type=snps;BUDDY=0 GT:AD:DP:GQ:PL ./.:1,1:1:1:1,1,1
Het call
chr2 48018081 . A G 851.77 PASS AC=1;AF=0.5;AN=2;ClippingRankSum=0;DP=123;ExcessHet=3.0103;FS=0;MLEAC=1;MLEAF=0.5;MQ=60;MQRankSum=0;set=Intersection;type=snps;BUDDY=0 GT:AD:DP:GQ:PL 0/1:29,33:62:99:880,0,869