Normalizing drug prefixes to DrugCentral

[caufieldjh]

Drugs with the following prefixes need to be mapped to their corresponding DrugCentral CURIEs:

• CHEBI (not ingested as a Drug, so keep the original and make an Association to DrugCentral instance)
• CHEMBL.COMPOUND
• DRUGBANK
• pharmgkb.drug (note this is currently ingested as PHARMGKB, just using a more specific prefix here)
• ttd.drug

Of these, PharmGKB and TTD do not provide direct maps to DrugCentral or vice-versa; maps will be curated based on shared mappings and exact string matching.

[caufieldjh]

Drugs ingested through PharmGKB and assigned CHEBI ids still get biolink:Drug so these should be normalized too

[caufieldjh]

CHEBI ids loaded from the ontology aren't normalized to DrugCentral, but each of them with a mapping to DrugCentral gets a new biolink:exact_match relation added to the chebi_edges.tsv to capture the mapping anyway.

[caufieldjh]

Realized that the CHEBI transform should also add the mapped DrugCentral nodes to the chebi_nodes.tsv, so that's next

[sonarcloud[bot]]

SonarCloud Quality Gate failed.   

0 Bugs
0 Vulnerabilities
0 Security Hotspots
3 Code Smells

26.7% Coverage
0.0% Duplication