'Technetium Tc-99m Albumin Aggregated' is split into two cliques

[amykglen]

we noticed that the SRI node normalizer appears to miss merging CHEMBL.COMPOUND:CHEMBL1201522 with the MESH:D013668 clique, though as far as we can tell those concepts are equivalent.

https://nodenormalization-sri.renci.org/get_normalized_nodes?curie=MESH:D013668

{
   "MESH:D013668":{
      "id":{
         "identifier":"MESH:D013668",
         "label":"Technetium Tc 99m Aggregated Albumin"
      },
      "equivalent_identifiers":[
         {
            "identifier":"MESH:D013668",
            "label":"Technetium Tc 99m Aggregated Albumin"
         },
         {
            "identifier":"UMLS:C0039415",
            "label":"Technetium Tc 99m Aggregated Albumin"
         },
         {
            "identifier":"DRUGBANK:DB09447"
         }
      ],
      "type":[
         "biolink:ChemicalEntity",
         "biolink:NamedThing",
         "biolink:PhysicalEssence",
         "biolink:ChemicalOrDrugOrTreatment",
         "biolink:ChemicalEntityOrGeneOrGeneProduct",
         "biolink:ChemicalEntityOrProteinOrPolypeptide",
         "biolink:PhysicalEssenceOrOccurrent"
      ]
   }
}

https://nodenormalization-sri.renci.org/get_normalized_nodes?curie=CHEMBL.COMPOUND:CHEMBL1201522

{
   "CHEMBL.COMPOUND:CHEMBL1201522":{
      "id":{
         "identifier":"CHEMBL.COMPOUND:CHEMBL1201522",
         "label":"TECHNETIUM TC 99M ALBUMIN AGGREGATED"
      },
      "equivalent_identifiers":[
         {
            "identifier":"CHEMBL.COMPOUND:CHEMBL1201522",
            "label":"TECHNETIUM TC 99M ALBUMIN AGGREGATED"
         }
      ],
      "type":[
         "biolink:ChemicalEntity",
         "biolink:NamedThing",
         "biolink:PhysicalEssence",
         "biolink:ChemicalOrDrugOrTreatment",
         "biolink:ChemicalEntityOrGeneOrGeneProduct",
         "biolink:ChemicalEntityOrProteinOrPolypeptide",
         "biolink:PhysicalEssenceOrOccurrent"
      ]
   }
}

just thought we would report this for future improvements!

[amykglen]

I suppose it's also worth noting that using drug_chemical_conflate=true doesn't resolve this one :)

[gaurav]

Thanks for reporting this, Amy! We now have manual concords for chemicals, so I'll add this there, so it'll work with drug_chemical_conflated=true in the Nov 8 release. But I'll leave this issue open so we try to work out how to fix this systematically.