Ingestion of VRS-annotated VCF is producing conflicting allele when state == "" #221
Description
Describe the bug
I noticed this during ingestion of a VCF that had previously been annotated with VRS IDs/attributes. Part of the ingestion process involves attempting to recreate the original allele from the given parameters. For VCF variants where the annotated VRS alt state is just a deletion ("."), the allele ID for the recreated VRS object does not match the given ID.
Steps to reproduce
For example, the ALT from this record
chrY 2840370 . C CGAGAGAGA . . AC=12;AC_Hemi=0;AC_Het=0;AC_Hom=12;AF=0.0165975;AN=723;VRS_Allele_IDs=ga4gh:VA.DjClXadsTSlpvZwfdqm-kwO78vasYuy4,ga4gh:VA.lMn5xCVH1wH_6CFGngryPHVkZq0IluGH;VRS_Starts=2840369,2840370;VRS_Ends=2840370,2840416;VRS_States=C,.
Expected behavior
The ID should be ga4gh:VA.lMn5xCVH1wH_6CFGngryPHVkZq0IluGH
Current behavior
Our code produces this allele (note the differing ID)
{'digest': 'xWSEycpHNOuAM23ZO9Fg_69uH_XAAyoC',
'id': 'ga4gh:VA.xWSEycpHNOuAM23ZO9Fg_69uH_XAAyoC',
'location': {'digest': 'SJwjlpriSSV9CRgpOnVtQzZfJkZ0Jsdu',
'end': 2977218,
'sequenceReference': {'refgetAccession': 'SQ.8_liLu1aycC0tPQPFmUaGXJLDs5SbPZ5',
'type': 'SequenceReference'},
'start': 2977180,
'type': 'SequenceLocation'},
'state': {'length': 0,
'repeatSubunitLength': 38,
'sequence': '',
'type': 'ReferenceLengthExpression'},
'type': 'Allele'}
Acceptance Criteria
that shouldn't happen
Possible reason(s)
No response
Suggested fix
No response
Branch, commit, and/or version
main
Screenshots
No response
Environment details
main
Additional details
At first glance, I think the provided VRS coordinates/state look pretty weird given the ALT
Contribution
None