Hi there, my data QC group is trying to determine whether the creators of our OMOP database set up the genomic CDM (https://github.com/OHDSI/Genomic-CDM) correctly. From looking at the schema for the genomic CDM it seems to me that the genomic_test table should contain one record for each given test type and linking to all genes that are tested for in this test in the target_gene table. There shouldn’t be a link back from these tables to the specimen table. The variant_occurrence table should then link variances found for a specific target_gene in a given specimen, which in turn links to a person. It seems, however, that our database was set up that each genomic_test isn’t a type of test with a unique name and version, but refers to a test given to a specific person, and that it only contains target_genes for which variants were found, making it impossible for us to query for gene mutations that we can exclude for a specific person/tumor. If anyone could comment of the intention of the genomic_test and target_gene tables in this context we would greatly appreciate it.
