Investigate the potential of using OMOP relationships to break down a precomposed assay term

@ibraun asked in the last call: "Here's an example of a LOINC term, https://athena.ohdsi.org/search-terms/terms/3031151, which for example uses a relationship ID Has system to state that it's using Body Fluid. My question was asking if we wanted to use relationship (not strictly subclass relationships) like this to pull out the individual aspects of a given assay after it had been concatenated (if necessary) from multiple fields and mapped to a single OMOP term, rather than keeping track of those individual fields that were used to map to the single OMOP term. It sounds like we can use those OMOP relationship IDs for this purpose, maybe in combination with the relationships provided by the mapping from that OMOP term to an OBO term, if I understood correctly (edited).

This is a very good question and worth investigating.

Not sure if we can prioritise this right this moment, but we should look into this after @james made a larger dent into the modelling.

• Try to find single example of assay terms with useful constituent relationships in the OMOP relationship table and report here.