make LOINC components standard

[dimshitc]

I have this source code "t(9;22)(q34.1;q11)(ABL1,BCR)" and LOINC has these more specific measurements t(9;22)(q34.1;q11)(ABL1,BCR) fusion transcript in Blood or Tissue by FISH t(9;22)(q34.1;q11)(ABL1,BCR) fusion transcript in Bone marrow by Molecular genetics method t(9;22)(q34.1;q11)(ABL1,BCR) fusion transcript in Blood or Tissue by Molecular genetics method etc, while the source data doesn't provide such a details. This way, the perfect mapping candidate is the LOINC Component concept t(9;22)(q34.1;q11)(ABL1,BCR) that stands exactly for this type of measurement So such LOINC Component concepts should be made standard. Moreover they will serve as ancestors, if I want to find all "T(9;22) ..." tests regardless of method, I would use t(9;22)(q34.1;q11)(ABL1,BCR) as an ancestor concept

[Alexdavv]

The funny thing is that translocations being the genetic abnormalities should reside in the OMOP Genomic vocabulary.

But for the rest, we considered LOINC Hierarchies (like that), not Components to be promoted to the Standard concepts. Components are too general and you never know where and how the thing is measured.

But even when you look at these Hierarchies, they don't explicitly say you what exactly is measured. Presence of the actual component, absolute amount in the whole specimen or in specific volume, or relative amount, susceptibility to something or again the actual concentration - everything rolls up to the same Hierarchies.

@p-talapova @MariaRohozhkina @TinyRickC137

[cgreich]

Let's stay in the problem space for a little, @dimshitc.

What is the analytic use case to distinguish t(9;22)(q34.1;q11)(ABL1,BCR) in Blood or Tissue by FISH, in Blood or Tissue by Molecular genetics method or in bone marrow by Molecular genetics method? You really want to filter that way? My hunch is you will be overly happy, even elated, to find such translocation information AT ALL.

[dimshitc]

That's exactly my point, @cgreich I don't want to distinguish between all different methods, that's way I'm asking to make the component standard, so I can have the translocation regardless of method. @Alexdavv here is a hypothetical example of a research question when I would need a component: I want to look at all the cases related to Methicillin resistant Staphylococcus aureus, maybe I want to create a feature like this. So I would need this code as a parent https://athena.ohdsi.org/search-terms/terms/1012565

I understand that this case is very hypothetical, so for now I need only the Genetic cases. What is the current process of enriching the OMOP genomic vocabulary? Is it the vocabulary contribution by the Oncology WG?

[cgreich]

all the cases related to Methicillin resistant Staphylococcus aureus

Answering only for translocations: We are adding them as OMOP Genomic ones. Without a method. You will be good.