[biomed nl] update embedding index with linked props examples (#4052)

deploy/nl/embeddings.yaml

@@ -1,4 +1,4 @@
 medium_ft: embeddings_medium_2024_03_14_16_38_53.ft_final_v20230717230459.all-MiniLM-L6-v2.csv
 sdg_ft: embeddings_sdg_2023_12_26_10_03_03.ft_final_v20230717230459.all-MiniLM-L6-v2.csv
 undata_ft: embeddings_undata_2024_03_20_11_01_12.ft_final_v20230717230459.all-MiniLM-L6-v2.csv
-bio_ft: embeddings_bio_2024_03_04_10_28_51.ft_final_v20230717230459.all-MiniLM-L6-v2.csv
+bio_ft: embeddings_bio_2024_03_19_16_39_03.ft_final_v20230717230459.all-MiniLM-L6-v2.csv

server/integration_tests/test_data/detection_api_bio/debug_info.json

@@ -16,6 +16,7 @@
     "PROP": [
       "phylum",
       "chemblID",
+      "virusGenus",
       "geneID",
       "geneticVariantFunctionalCategory",
       "typeOfGene",
@@ -28,15 +29,13 @@
       "hg19GenomicLocation",
       "strandOrientation",
       "alleleOrigin",
-      "referenceSNPClusterID<-GeneticVariantGeneAssociation->geneSymbol",
+      "<-referenceSNPClusterID{typeOf:GeneticVariantGeneAssociation}->geneSymbol",
       "ofVirusSpecies",
       "antigenType",
       "chromosomeSize",
       "ncbiTaxonID",
       "simplifiedMolecularInputLineEntrySystem",
-      "ncbiProteinAccessionNumber",
-      "virusHost",
-      "imageUrl"
+      "ncbiProteinAccessionNumber"
     ]
   }
 }

tools/nl/embeddings/data/curated_input/bio/README.md

@@ -0,0 +1,12 @@
+# Curated Input for Bio index
+
+This index has properties used by biomedical entities and follows the format of [relation expressions](https://docs.datacommons.org/api/rest/v2#relation-expressions). Properties can be structured like:
+
+- `prop`: this can match to either in or out properties
+    - e.g., `virusHost` which will match both the 'in' and 'out' values for the property virusHost
+- `->prop`: this matches to an 'out' property
+    - e.g., `->phylum` which will match the 'out' values for the property phylum
+- `<-prop`: this matches to an 'in' property
+    - e.g., `<-virusGenus` which will match the 'in' values for the property virusGenus
+- `<-prop1{typeOf:X}->prop2`: in this case, we will get all the 'in' values for prop1 that are of type X & then from those values, get all the 'out' values for prop2
+    - e.g., `<-geneID{typeOf:DiseaseGeneAssociation}->diseaseOntologyID` which will first get all the DiseaseGeneAssociations that are 'in' values for the property geneID and then get all the 'out' values for the property diseaseOntologyID for those DiseaseGeneAssociations.

tools/nl/embeddings/data/curated_input/bio/sheets_svs.csv

@@ -1,45 +1,48 @@
 dcid,Name,Description,Override_Alternatives,Curated_Alternatives
-ofVirusSpecies,ofVirusSpecies,"The species of a virus isolate",,
-virusHost,virusHost,"A specific organism or taxonomic group of organisms that are susceptible to be infected by a virus",,"host of a virus"
-ncbiTaxonID,ncbiTaxonID,"NCBI Taxonomy database identifier",,
-diseaseName,diseaseName,"preferred disease name for the concept specified by disease identifiers",,"The name of the disease"
-observedAllele,observedAllele,"The sequences of the observed alleles from rs-fasta files.",,
-referenceAlleleNCBI,referenceAlleleNCBI,"Reference genomic sequence from dbSNP",,"reference allele"
+ofVirusSpecies,ofVirusSpecies,The species of a virus isolate,,
+virusHost,virusHost,A specific organism or taxonomic group of organisms that are susceptible to be infected by a virus,,host of a virus
+ncbiTaxonID,ncbiTaxonID,NCBI Taxonomy database identifier,,
+diseaseName,diseaseName,preferred disease name for the concept specified by disease identifiers,,The name of the disease
+observedAllele,observedAllele,The sequences of the observed alleles from rs-fasta files.,,
+referenceAlleleNCBI,referenceAlleleNCBI,Reference genomic sequence from dbSNP,,reference allele
 class,class,,,
 phylum,phylum,,,
-geneticVariantFunctionalCategory,geneticVariantFunctionalCategory,"Functional category of the genetic variant",,
-hg19GenomicPosition,hg19GenomicPosition,"The genomic position of a genetic variant using the hg19 assembly",,
-hg19GenomicLocation,hg19GenomicLocation,"The genomic location of a genetic variant using the hg19 assembly",,
-hg38GenomicPosition,hg38GenomicPosition,"The genomic position of a genetic variant using the hg38 assembly",,
-hg38GenomicLocation,hg38GenomicLocation,"The genomic location of a genetic variant using the hg38 assembly",,
-hasRNATranscript,hasRNATranscript,"Recorded transcript",,"RNA transcript that a gene has"
-strandOrientation,strandOrientation,"The strand on which a given annotation is located",,"The orientation of the strand on which an annotation is located"
-typeOfGene,typeOfGene,"The type of gene",,
-omimID,omimID,"OMIM database identifier",,
+geneticVariantFunctionalCategory,geneticVariantFunctionalCategory,Functional category of the genetic variant,,
+hg19GenomicPosition,hg19GenomicPosition,The genomic position of a genetic variant using the hg19 assembly,,
+hg19GenomicLocation,hg19GenomicLocation,The genomic location of a genetic variant using the hg19 assembly,,
+hg38GenomicPosition,hg38GenomicPosition,The genomic position of a genetic variant using the hg38 assembly,,
+hg38GenomicLocation,hg38GenomicLocation,The genomic location of a genetic variant using the hg38 assembly,,
+hasRNATranscript,hasRNATranscript,Recorded transcript,,RNA transcript that a gene has
+strandOrientation,strandOrientation,The strand on which a given annotation is located,,The orientation of the strand on which an annotation is located
+typeOfGene,typeOfGene,The type of gene,,
+omimID,omimID,OMIM database identifier,,
 icd10CMCode,icd10CMCode,"The disease diagnosis code for version 10 of the International Classification of Diseases (ICD), Clinical Modification",,
-subClassificationOf,subClassificationOf,"subclassification of",,
-snomedCT,snomedCT,"Systematiized Nomenclature of Medicine (SNOMED) clinical terms (CT) code",,
-unifiedMedicalLanguageSystemConceptUniqueIdentifier,unifiedMedicalLanguageSystemConceptUniqueIdentifier,"Unified Medical Language System (UMLS) Concept Unique Identifier (CUI)",, "UMLS CUI"
-specializationOf,specializationOf,"specialization of",,
-chemblID,chemblID,"ChEMBL identifier",,
-simplifiedMolecularInputLineEntrySystem,simplifiedMolecularInputLineEntrySystem,"Simplified Molecular Input Line Entry System (SMILE)",,
-medicalSubjectHeadingSupplementaryRecordID,medicalSubjectHeadingSupplementaryRecordID,"A unique ID for a Medical Subject Heading supplementary record",,"An ID for a Medical Subject Heading supplementary record;MeSH supplementary record ID"
-medicalSubjectHeadingDescriptorID,medicalSubjectHeadingDescriptorID,"A unique ID for a Medical Subject Heading Descriptor record",,"An ID for a Medical Subject Heading descriptor record;MeSH descriptor record ID"
+subClassificationOf,subClassificationOf,subclassification of,,
+snomedCT,snomedCT,Systematiized Nomenclature of Medicine (SNOMED) clinical terms (CT) code,,
+unifiedMedicalLanguageSystemConceptUniqueIdentifier,unifiedMedicalLanguageSystemConceptUniqueIdentifier,Unified Medical Language System (UMLS) Concept Unique Identifier (CUI),," ""UMLS CUI"""
+specializationOf,specializationOf,specialization of,,
+chemblID,chemblID,ChEMBL identifier,,
+simplifiedMolecularInputLineEntrySystem,simplifiedMolecularInputLineEntrySystem,Simplified Molecular Input Line Entry System (SMILE),,
+medicalSubjectHeadingSupplementaryRecordID,medicalSubjectHeadingSupplementaryRecordID,A unique ID for a Medical Subject Heading supplementary record,,An ID for a Medical Subject Heading supplementary record;MeSH supplementary record ID
+medicalSubjectHeadingDescriptorID,medicalSubjectHeadingDescriptorID,A unique ID for a Medical Subject Heading Descriptor record,,An ID for a Medical Subject Heading descriptor record;MeSH descriptor record ID
 activeIngredient,activeIngredient,"component that provides pharmacological activity or other direct effect in the diagnosis, cure, mitigation, treatment, or prevention of disease, or to affect the structure or any function of the body of man or animals",,
-administrationRoute,administrationRoute,"The method by which a drug is administered",,
-dosageForm,dosageForm,"physical form in which a drug is produced and dispensed",,
-antibodyType,antibodyType,"type of antibody",,
-antigenType,antigenType,"type of antigen",,
-chromosomeSize,chromosomeSize,"number of nucleotides in a chromosome",,"Size of chromosome"
-ensemblID,ensemblID,"Ensembl ID",,
-fullName,fullName,"full name of the gene",,
-geneID,geneID,"gene id",,
-ncbiProteinAccessionNumber,ncbiProteinAccessionNumber,"NCBI protein accession number",,
-alleleOrigin,alleleOrigin,"Variant allele origin",,"Origin of variant allele"
-alleleType,alleleType,"The allele of a genetic variant observed within a population",,"Type of allele"
-ncbiDNASequenceName,ncbiDNASequenceName,"NCBI defined segment of DNA sequence name",,"Name used by NIH NCBI to refer to a segment of DNA sequence"
-imageUrl,imageUrl,"url to an image of what the biological specimen looks like",,"what the entity looks like"
-genomicCoordinates,genomicCoordinates,"genomic coordinates",,
-availableStrength,availableStrength,"dose approved for a drug",,
-referenceSNPClusterID<-GeneticVariantGeneAssociation->geneSymbol,GeneticVariantGeneAssociation,"Association between a genetic variant and a gene",,"Gene associated with a genetic variant;genetic variant associated with a gene"
-diseaseOntologyID<-DiseaseGeneAssociation->geneID,DiseaseGeneAssociation,"Association of a disease and a gene",,
+administrationRoute,administrationRoute,The method by which a drug is administered,,
+dosageForm,dosageForm,physical form in which a drug is produced and dispensed,,
+antibodyType,antibodyType,type of antibody,,
+antigenType,antigenType,type of antigen,,
+chromosomeSize,chromosomeSize,number of nucleotides in a chromosome,,Size of chromosome
+ensemblID,ensemblID,Ensembl ID,,
+fullName,fullName,full name of the gene,,
+geneID,geneID,gene id,,
+ncbiProteinAccessionNumber,ncbiProteinAccessionNumber,NCBI protein accession number,,
+alleleOrigin,alleleOrigin,Variant allele origin,,Origin of variant allele
+alleleType,alleleType,The allele of a genetic variant observed within a population,,Type of allele
+ncbiDNASequenceName,ncbiDNASequenceName,NCBI defined segment of DNA sequence name,,Name used by NIH NCBI to refer to a segment of DNA sequence
+imageUrl,imageUrl,url to an image of what the biological specimen looks like,,what the entity looks like
+genomicCoordinates,genomicCoordinates,genomic coordinates,,
+availableStrength,availableStrength,dose approved for a drug,,
+<-referenceSNPClusterID{typeOf:GeneticVariantGeneAssociation}->geneSymbol,GeneticVariantGeneAssociation,Gene associated with a genetic variant,,
+<-geneSymbol{typeOf:GeneticVariantGeneAssociation}->referenceSNPClusterID,GeneticVariantGeneAssociation,genetic variant associated with a gene,,
+<-diseaseOntologyID{typeOf:DiseaseGeneAssociation}->geneID,DiseaseGeneAssociation,Gene associated with a disease,,
+<-geneID{typeOf:DiseaseGeneAssociation}->diseaseOntologyID,DiseaseGeneAssociation,Disease associated with a gene,,
+virusGenus,virusGenus,genus of a virus species,,

tools/nl/embeddings/data/preindex/bio/duplicate_names.csv

@@ -1,4 +1,6 @@
 PreferredSV,DroppedSV,DuplicateName
+<-referenceSNPClusterID{typeOf:GeneticVariantGeneAssociation}->geneSymbol,<-geneSymbol{typeOf:GeneticVariantGeneAssociation}->referenceSNPClusterID,GeneticVariantGeneAssociation
+<-diseaseOntologyID{typeOf:DiseaseGeneAssociation}->geneID,<-geneID{typeOf:DiseaseGeneAssociation}->diseaseOntologyID,DiseaseGeneAssociation
 Amount_EconomicActivity_GrossDomesticProduction_Nominal,dc/topic/GDP,GDP
 dc/topic/Mortality,dc/topic/WHOMortality,Mortality
 dc/topic/EconomicActivity,dc/topic/GlobalEconomicActivity,Economic Activity