The QUAL column generated in the VCF files by CANVAS follows the standard rule found in The Variant Call Format (VCF) Version 4.2 Specification for that column. In other words, for CANVAS the QUAL columns mean this?
- Phred-scaled quality score for the assertion made in ALT. i.e. −10log10 prob(call in ALT is wrong). If ALT is ‘.’ (no variant) then this is −10log10 prob(variant), and if ALT is not ‘.’ this is −10log10 prob(no variant). If unknown, the missing value should be specified. (Numeric)*
I am asking this question because I am checking the quality of the CNVs called by this program in the 100K genomes project. and most of these variants present a QUAL less than 10. Most of them don't get the PASS filter.
For example, for a complete genome, canvas calls 411 CNVs but only 59 got the PASS filter. Looking at the data, these 59 present a on average a QUAL = 16.65 The variants that didn't pass all filters their QUAL is 2.91 on average.
As a bioinformatician recently working in a clinical lab, this data is deficient. Am I biased? I am missing something here??
Thanks!
