Some long_table.json files contain variants where the Gene field includes multiple genes separated by & (e.g. GENE1&GENE2).
The current database update process does not handle these cases correctly.
As an immediate solution that does not require changes to the annotation pipeline, update the update-db module so that when a
Gene field contains &, it splits the value into individual genes before importing the variants into the database.
Considerations
Before implementing this, it is important to verify how annotations differ between:
- The original fusion gene annotation (e.g. GENE1&GENE2)
- The annotations obtained when each gene is considered independently
Initial support could store the variant using the existing fusion gene annotation, but the preferred long-term approach would be to retrieve and store the individual annotations for each gene separately, as these may contain different and more accurate information.
Some
long_table.jsonfiles contain variants where the Gene field includes multiple genes separated by&(e.g.GENE1&GENE2).The current database update process does not handle these cases correctly.
As an immediate solution that does not require changes to the annotation pipeline, update the
update-dbmodule so that when aGenefield contains&, it splits the value into individual genes before importing the variants into the database.Considerations
Before implementing this, it is important to verify how annotations differ between:
Initial support could store the variant using the existing fusion gene annotation, but the preferred long-term approach would be to retrieve and store the individual annotations for each gene separately, as these may contain different and more accurate information.