Grounding records have extraineous fields

[jvwong]

There are cases where a participant (entity) association record has extra fields. These fields are unrelated to the type, and appear to be merged from another record.

Example:

• Document Sci Signal 13 2020](https://biofactoid.org/document/034d7fcc-80a0-4b78-8b1d-4fd8652dc4f8)
• Entity
  • ID: d8110751-0743-4354-8986-e05968905db2
  • Name: CtBP Monomer

Association (extra fields like Organism - looks like record from poly(rC) binding protein 2)

"association": {
    "charge": 0 ,
    "combinedOrganismIndex": 0 ,
    "dbName": "ChEBI" ,
    "dbPrefix": "CHEBI" ,
    "dbXrefs": [
        {
            "db": "MGI" ,
            "id": "MGI:108202"
        } ,
        {
            "db": "Ensembl" ,
            "id": "ENSMUSG00000056851"
        }
    ] ,
    "defaultOrganismIndex": 0 ,
    "distance": 0.6842105263157895 ,
    "esScore": 1.3025038 ,
    "formulae": [
        "C14H23N3OS"
    ] ,
    "id": "10056" ,
    "inchi": "InChI=1S/C14H23N3OS/c1-3-4-5-6-10-18-14-13(15-19-16-14)12-8-7-9-17(2)11-12/h8H,3-7,9-11H2,1-2H3" ,
    "inchiKey": "JOLJIIDDOBNFHW-UHFFFAOYSA-N" ,
    "mass": 281.417 ,
    "monoisotopicMass": 281.15618 ,
    "name": "xanomeline" ,
    "nameDistance": 0.6842105263157895 ,
    "namespace": "chebi" ,
    "organism": "10090" ,
    "organismIndex": 0 ,
    "organismName": "Mus musculus" ,
    "overallDistance": 6800000068 ,
    "shortSynonyms": [
        "xanomeline" ,
        "LY 246708" ,
        "5-[4-(hexyloxy)-1,2,5-thiadiazol-3-yl]-1-methyl-1,2,3,6-tetrahydropyridine"
    ] ,
    "summary": "A canonical ribonucleotide residue derived from cytidine 5'-monophosphate" ,
    "synonyms": [
        "LY 246708" ,
        "5-[4-(hexyloxy)-1,2,5-thiadiazol-3-yl]-1-methyl-1,2,3,6-tetrahydropyridine"
    ] ,
    "type": "chemical"
}

here's what comes out of the grounding service for CHEBI:10056:

{
	"namespace": "chebi",
	"type": "chemical",
	"dbName": "ChEBI",
	"dbPrefix": "CHEBI",
	"id": "10056",
	"name": "xanomeline",
	"inchi": "InChI=1S/C14H23N3OS/c1-3-4-5-6-10-18-14-13(15-19-16-14)12-8-7-9-17(2)11-12/h8H,3-7,9-11H2,1-2H3",
	"inchiKey": "JOLJIIDDOBNFHW-UHFFFAOYSA-N",
	"synonyms": [
		"LY 246708",
		"5-[4-(hexyloxy)-1,2,5-thiadiazol-3-yl]-1-methyl-1,2,3,6-tetrahydropyridine"
	],
	"charge": 0,
	"mass": 281.417,
	"monoisotopicMass": 281.15618,
	"formulae": [
		"C14H23N3OS"
	]
}

[jvwong]

A better example:

• Front Immunol 9 2018](https://biofactoid.org/document/e49f3cd8-5cc5-497b-a5be-d7588a65b80a)
• Entity
  • ID: a18f45f7-8c9e-4553-b527-d7f40647e105
  • Name: Butyrate

• association:

"association": {
    "charge": -1 ,
    "combinedOrganismIndex": 0 ,
    "dbName": "ChEBI" ,
    "dbPrefix": "CHEBI" ,
...
    "namespace": "chebi" ,
    "organism": "511145" ,
    "organismIndex": 0 ,
    "organismName": "Escherichia coli" ,
    "overallDistance": 1000 , 
...

[jvwong]

I think this bug may have been squashed:

• Failing case for paper: No related papers; grounding issue #965
• Re-association of entity removes out existing fields #1005

and that these documents represent remnants of the bug that preceded the fix.

Double check this.

[jvwong]

A pathway with chemical (Ca) with what appears to be an erroneous 'Summary' field from cytidine 5'-monophosphate residue:

• document PLoS Biol 18 2020
• Entity
  • Name: 'calcium'
  • Id: c7885c87-d91e-48d0-80b3-4af2cdc8b08a
  • association

"association": {
...
    "dbName": "ChEBI" ,
    "dbPrefix": "CHEBI" ,
    "defaultOrganismIndex": 0 ,
...
    "summary": "A canonical ribonucleotide residue derived from cytidine 5'-monophosphate" ,
    "synonyms": [
        "Ca(0)" ,
        "Can" ,
        "calcium"
    ] ,
    "type": "chemical"
} ,