Genetic Condition
Requirements:
- Should genetic conditions transfer across curations?
- If yes, we will need a system for managing genetic conditions.
- If no, the requirements for querying the data to bring together related curations may become more complicated, but the requirements on implementation are lighter.
- Not necessarily an either-or; there may be a requirement to support both unregistered genetic conditions as well as a genetic condition registry.
Solution w/o registry
- Identify when a genetic condition exsits that meets the requirements from the user. This could mean:
- The user has specified a genetic condition target already for the condition (either an OMIM id with a gene target, or a previously specified GC.
- In the case that the user has specified a GC, return that identifier, leave message unmodified.
- In the case that the user did not directly specify a GC, restructure the message with one:
- Make the target of the association a blank node:
- subClassOf the previous condition
- associated gene linked
- Label is
,
Queries involving unregistered conditions
- Likely need a query to find genetic conditions beneath the containing condition.
- Can be part of the GraphQL schema
- Need a type for genetic condition
Minting a new genetic condition
Modifying existing message
vs. Genetic Disease Registry
vs. Regular Disease Registry
IRI
Options are:
- Treat as blank node, no IRI
- This is a problem if multiple curations need to reference the same disorder; blank nodes really only have significance within a single document,