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
Truvari doesn't consider ALT=* alleles and filters them the same a reference monozygotic (ALT=.) variants.
In section 1.6.1 of the VCF v4.5 specification, we see inside the definition for ALT that the ‘*’ symbol (allele missing due to overlapping deletion). My understanding is '' represents an allele which cannot be present and therefore isn't a variant per-say but more of an annotation that a variant call isn't possible (i.e. a 'null' variant). As a hypothetical example, the below diploid alignment would create a '' allele.
Hap1 - ATA---ATA
Hap2 - ATACTCATA
REF - ATACGCATA
CHROM POS ID REF ALT ... GT
chr 3 . ACGC A ... 1|0
chr 5 . G T,* ... 2|1
These *-alleles would create redundancies for counting the variant states (TP/FP/FN) as its state is always equal to its overlapping deletion's state.
It gets even more complicated if we consider the possibilities of the SNP at position 5 as a TP or FP. Had the caller not made that SNP, the asterisks allele wouldn't be present and therefore we wouldn't be counted as a TP/FP (depending on the deletion's state).
This complication also holds if the above example is the baseline VCF and the caller did not report the SNP (a FN). In this case the asterisks allele must become a FN regardless of if its overlapping deletion is TP or FN.
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
-
Truvari doesn't consider
ALT=*
alleles and filters them the same a reference monozygotic (ALT=.
) variants.In section 1.6.1 of the VCF v4.5 specification, we see inside the definition for ALT that
the ‘*’ symbol (allele missing due to overlapping deletion)
. My understanding is '' represents an allele which cannot be present and therefore isn't a variant per-say but more of an annotation that a variant call isn't possible (i.e. a 'null' variant). As a hypothetical example, the below diploid alignment would create a '' allele.These *-alleles would create redundancies for counting the variant states (TP/FP/FN) as its state is always equal to its overlapping deletion's state.
It gets even more complicated if we consider the possibilities of the SNP at position 5 as a TP or FP. Had the caller not made that SNP, the asterisks allele wouldn't be present and therefore we wouldn't be counted as a TP/FP (depending on the deletion's state).
This complication also holds if the above example is the baseline VCF and the caller did not report the SNP (a FN). In this case the asterisks allele must become a FN regardless of if its overlapping deletion is TP or FN.
Beta Was this translation helpful? Give feedback.
All reactions