Describe the bug
Our PGx annotation for rs72549309 (ATGA/del genotype) is not correctly linking to the DPYD gene.
There are 2127 evidence (~6%) with a null target and I expect that most of them fall under the same scenario.
Observed behaviour
This PGx evidence is missing gene annotation that is available in PharmGKB.
-RECORD 0------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
datasourceId | pharmgkb
datasourceVersion | 2024-01-05
datatypeId | clinical_annotation
directionality | NULL
drugFromSource | fluorouracil
drugFromSourceId | CHEBI_46345
evidenceLevel | 1A
genotype | ATGA/del
genotypeAnnotationText | The del allele of rs72549309 is assigned no function by CPIC. Patients with the ATGA/del genotype may have decreased DPYD activity as compared to those with the ATGA/ATGA genotype. Other genetic and clinical factors may also influence catalytic activity of DPYD.
genotypeId | NULL
haplotypeFromSourceId | NULL
haplotypeId | NULL
literature | [10071185, 24648345]
pgxCategory | other
phenotypeFromSourceId | NULL
phenotypeText | NULL
studyId | 1447989678
targetFromSourceId | NULL
variantFunctionalConsequenceId | NULL
variantRsId | rs72549309
Expected behaviour
This is a similar case to the one in #3204, only that I think that here VEP is not returning a gene annotation because the genotype is null. The solution is the same, if we can't have the gene from VEP, we'll take the one reported by PGKB which is present for the vast majority of cases.
This implementation will be done by the EVA team, with a prior analysis of significant discrepancies between VEP and PharmGKB genes. We could have a new version for this release, but it is uncertain.
No action on our side.
Describe the bug Our PGx annotation for rs72549309 (ATGA/del genotype) is not correctly linking to the DPYD gene.
There are 2127 evidence (~6%) with a null target and I expect that most of them fall under the same scenario.
Observed behaviour This PGx evidence is missing gene annotation that is available in PharmGKB.
Expected behaviour This is a similar case to the one in #3204, only that I think that here VEP is not returning a gene annotation because the genotype is null. The solution is the same, if we can't have the gene from VEP, we'll take the one reported by PGKB which is present for the vast majority of cases.
This implementation will be done by the EVA team, with a prior analysis of significant discrepancies between VEP and PharmGKB genes. We could have a new version for this release, but it is uncertain. No action on our side.