VAF considered by HAPPY sensitivity and specificity

[Belle-AM]

We are using hap.py to compare various sequencing runs processed via the DRAGEN pipeline. In the DRAGEN output vcf, any variant with a VAF of at least 0.01 is reported. However, any variant with less than a VAF of 0.05 is marked as lowvaf in its filter column.

When hap.py reportsthe "PASS" results for concordance, I assume that the comparison excludes variants with the lowvaf filter, and thus sensitivity and specificity are based on only those variants with at least 0.05 VAF. Can this be confirmed?