How to get variant allele frequency for internal tandem duplications from Pindel?

myourshaw · 01-30-2020, 09:29 AM

I'm using the latest version of Pindel (0.2.5b9, 20160729) for somatic DNA to detect internal tandem duplications (ITDs) at variant allele frequencies (VAFs) in the range of 5% and greater. The vcf files produced by pindel2vcf always report 0 as the allele depth (AD) for the reference allele. Supposedly Pindel's VCFs also have a "reference depth" metric (RD) but this is not actually reported.

I cannot find anything in the raw Pindel output (such as the _TD file) that corresponds to an allele frequency or reference depth.

One (ugly) idea I have is to get total depth from something like GATK's DepthOfCoverage and divide that by Pindel's alt allele depth?

Is there a more elegant/correct solution to this issue?

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Variant allele frequency or allelic balance?	Kulvait	Bioinformatics	4	02-27-2019 09:22 AM
SSPACE creates tandem duplications instead of merging	seb.lees	Bioinformatics	0	05-28-2013 09:40 AM
how to compute Minor Allele Frequency for a variant call?	BhariD	Bioinformatics	0	05-07-2013 11:50 AM
how does varscan somatic compute variant allele frequency?	genomicist	Bioinformatics	0	06-11-2012 04:21 AM

01-30-2020, 09:29 AM	#1
myourshaw Junior Member Location: Los Angeles CA Join Date: Sep 2008 Posts: 2	How to get variant allele frequency for internal tandem duplications from Pindel? I'm using the latest version of Pindel (0.2.5b9, 20160729) for somatic DNA to detect internal tandem duplications (ITDs) at variant allele frequencies (VAFs) in the range of 5% and greater. The vcf files produced by pindel2vcf always report 0 as the allele depth (AD) for the reference allele. Supposedly Pindel's VCFs also have a "reference depth" metric (RD) but this is not actually reported. I cannot find anything in the raw Pindel output (such as the _TD file) that corresponds to an allele frequency or reference depth. One (ugly) idea I have is to get total depth from something like GATK's DepthOfCoverage and divide that by Pindel's alt allele depth? Is there a more elegant/correct solution to this issue?