You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I have some questions about the application of PON in GATK Mutect2. I have run some samples with and without PON (other parameters are exactly the same). The vcf.gz file from with-PON have less than 20K sites in chr22.
PON Tag in vcf.gz file
However, In the vcf.gz file I received from with-PON option, there are also some sites with PON tag? Why they are not removed? Looks to me, some sites in PON will be directly removed and not reported in the vcf.gz file, while some sites will be reported but contains PON tag (in below screenshot). Why is that?
So these are sites in active regions but also in PON. So I believe they should also be filtered out right?
is the PON filtering is allele-specific?
I found an question 5 years ago here. In this answer, it seems PON filtering in Mutect2 is NOT allele-specific. How is it now?
The logic behind PON filtering.
Third, I want to confirm the logic behind PON filtering. I believe it's only based on coordinates right? There is no statistic behind, if a site in variant calling is overlapped with a site in PON, it will be removed, or labelled as PON in the reported vcf.gz file.
The text was updated successfully, but these errors were encountered:
Hi GATK Team:
I have some questions about the application of PON in GATK Mutect2. I have run some samples with and without PON (other parameters are exactly the same). The vcf.gz file from with-PON have less than 20K sites in chr22.
PON Tag in vcf.gz file
However, In the vcf.gz file I received from with-PON option, there are also some sites with PON tag? Why they are not removed? Looks to me, some sites in PON will be directly removed and not reported in the vcf.gz file, while some sites will be reported but contains PON tag (in below screenshot). Why is that?
I think I found the explanation here:
So these are sites in active regions but also in PON. So I believe they should also be filtered out right?
is the PON filtering is allele-specific?
I found an question 5 years ago here. In this answer, it seems PON filtering in Mutect2 is NOT allele-specific. How is it now?
The logic behind PON filtering.
Third, I want to confirm the logic behind PON filtering. I believe it's only based on coordinates right? There is no statistic behind, if a site in variant calling is overlapped with a site in PON, it will be removed, or labelled as PON in the reported vcf.gz file.
The text was updated successfully, but these errors were encountered: