Merge pull request #128 from NCATS-Tangerine/publication-keywords

Publication keywords
NCATS-Tangerine · Aug 14, 2023 · 1137479 · 1137479
2 parents b08b9b3 + 16b13cd
commit 1137479
Show file tree

Hide file tree

Showing 18 changed files with 185 additions and 177 deletions.
diff --git a/CTD/smartapi.yaml b/CTD/smartapi.yaml
@@ -139,7 +139,7 @@ paths:
       - "$ref": "#/components/x-bte-kgs-operations/pathway2gene_2"
 components:
   x-bte-kgs-operations:
-    ## - pubmed IDs are |-delimited strings but BTE seems to parse these correctly
+    ## - pubmed IDs are pipe-delimited (|) strings but BTE seems to parse these correctly
     ## - while the API supports multiple IDs as input (batch-query)...we aren't using it because
     ##   response doesn't separate info from 1 input ID vs another. So BTE has trouble processing it
     ##   (would need a custom api-response-transform)
@@ -537,47 +537,47 @@ components:
   x-bte-response-mapping:
     chemical2gene:
       NCBIGene: data.GeneId            ## no prefix
-      pubmed: data.PubMedIds           ## no prefix
+      ref_pmid: data.PubMedIds             ## no prefix
       ## commenting out because data-processing / biolink-modeling issues
       # inTaxonName: data.Organism       ## human-readable label 
       # inTaxon: data.OrganismId         ## no prefix
     gene2chemical:
       MESH: data.ChemicalId            ## no prefix
-      pubmed: data.PubMedIds           
+      ref_pmid: data.PubMedIds           
       ## commenting out because data-processing / biolink-modeling issues
       # inTaxonName: data.Organism       ## human-readable label 
       # inTaxon: data.OrganismId
     # chemical2disease_1:
     #   MESH: data.DiseaseID          ## HAS prefix
     #   ctd_chemical_disease_interaction_types: data.DirectEvidence
-    #   pubmed: data.PubMedIDs
+    #   ref_pmid: data.PubMedIDs
     disease2chemical:
       MESH: data.ChemicalID
-      pubmed: data.PubMedIDs
+      ref_pmid: data.PubMedIDs
       ## commenting out because data-processing / biolink-modeling issues
       # ctd_chemical_disease_interaction_types: data.DirectEvidence  ## |-delimited string is kept right now
     # chemical2disease_2:
     #   OMIM: data.DiseaseID          ## HAS PREFIX
     #   ctd_chemical_disease_interaction_types: data.DirectEvidence
-    #   pubmed: data.PubMedIDs
+    #   ref_pmid: data.PubMedIDs
     chemical2go:
       GO: data.GoTermID  ## HAS PREFIX
       ## commenting out because data-processing / biolink-modeling issues
       # related_gene_symbols: data.GeneSymbol
       # related_gene_ncbigene: data.GeneID  ## no prefix
     disease2gene:
       NCBIGene: data.GeneID
-      pubmed: data.PubMedIDs
+      ref_pmid: data.PubMedIDs
       ## commenting out because data-processing / biolink-modeling issues
       # ctd_gene_disease_interaction_types: data.DirectEvidence
     # gene2disease_1:
     #   MESH: data.DiseaseID  ## HAS PREFIX
     #   ctd_gene_disease_interaction_types: data.DirectEvidence
-    #   pubmed: data.PubMedIDs
+    #   ref_pmid: data.PubMedIDs
     # gene2disease_2:
     #   OMIM: data.DiseaseID  ## HAS PREFIX
     #   ctd_gene_disease_interaction_types: data.DirectEvidence
-    #   pubmed: data.PubMedIDs
+    #   ref_pmid: data.PubMedIDs
     # gene2pathway_1:
     #   REACT: data.PathwayID  ## HAS PREFIX
     #   output_name: data.PathwayName

diff --git a/EBIgene2phenotype/smartapi.yaml b/EBIgene2phenotype/smartapi.yaml
@@ -623,8 +623,8 @@ components:
       response_mapping:
         "$ref": "#/components/x-bte-response-mapping/disease-object"
       # testExamples:
-      #   - qInput: "HGNC:10031"        ## RMRP
-      #     oneOutput: "OMIM:250250"    ## CARTILAGE-HAIR HYPOPLASIA
+      #   - qInput: "HGNC:1020"         ## BCS1L
+      #     oneOutput: "OMIM:603358"    ## GRACILE SYNDROME
     disease_to_gene:
     - supportBatch: true
       useTemplating: true
@@ -659,7 +659,7 @@ components:
     disease-object:
       OMIM: gene2phenotype.disease.disease_mim   ## no prefix
       output_name: gene2phenotype.disease.disease_name
-      pubmed: gene2phenotype.pmids   ## no prefix
+      ref_pmid: gene2phenotype.pmids   ## no prefix
       ## commenting out because data-processing / biolink-modeling issues
       # ## similar to "biolink:has_mode_of_inheritance" (predicate) but doesn't use the expected ontology terms and seems a bit different
       # g2p_allelic_requirement: gene2phenotype.allelic_requirement

diff --git a/MGIgene2phenotype/smartapi.yaml b/MGIgene2phenotype/smartapi.yaml
@@ -677,7 +677,7 @@ components:
       DOID: mgi.associated_with_disease.doid  ## has prefix
     gene_related_to_phenotype:
       MP: mgi.associated_with_phenotype.mp  ## has prefix
-      pubmed: mgi.associated_with_phenotype.pubmed  ## no prefix
+      ref_pmid: mgi.associated_with_phenotype.pubmed  ## no prefix
       ## commenting out because data-processing / biolink-modeling issues
       # mgi_allele_id: mgi.associated_with_phenotype.allele_id  ## MGI curie
       # mgi_allele_symbol: mgi.associated_with_phenotype.allele_symbol  ## free text?

diff --git a/bindingdb/smartapi.yaml b/bindingdb/smartapi.yaml
@@ -599,7 +599,7 @@ components:
         #     relation.curation_datasource,relation.pmid,relation.bindingdb_link,relation.article_doi,relation.patent_number
           fields: >-
             object.pubchem_cid,
-            subject.name,relation.pmid
+            subject.name,relation.pmid,relation.bindingdb_link,relation.article_doi
           size: 1000
         predicate: physically_interacts_with
         source: "infores:bindingdb"
@@ -629,7 +629,7 @@ components:
           #   relation.curation_datasource,relation.pmid,relation.bindingdb_link,relation.article_doi,relation.patent_number
           fields: >-
             subject.uniprot.accession,
-            subject.name,relation.pmid
+            subject.name,relation.pmid,relation.bindingdb_link,relation.article_doi
           size: 1000
         predicate: physically_interacts_with
         source: "infores:bindingdb"
@@ -639,23 +639,28 @@ components:
         #   - qInput: "PUBCHEM.COMPOUND:134553288"     ## US10736883, Example 387.0
         #     oneOutput: "UniProtKB:P35414"            ## Apelin receptor
   x-bte-response-mapping:
+  ## note on references:
+  ##   sometimes the doi + pmid will refer to the same publication (so..."duplicate" references)
+  ##   but there are plenty of cases where relationships have only 1 of these fields:
+  ##   only doi: https://biothings.transltr.io/bindingdb/query?q=_exists_:relation.article_doi%20AND%20(NOT%20_exists_:relation.pmid)
+  ##   only pmid: https://biothings.transltr.io/bindingdb/query?q=_exists_:relation.pmid%20AND%20(NOT%20_exists_:relation.article_doi)
     pubchem-object:
       "PUBCHEM.COMPOUND": object.pubchem_cid       ## no prefix
       "biolink:original_subject": subject.name     ## was called "Target Name Assigned by Curator or DataSource"
-      pubmed: relation.pmid
+      ref_pmid: relation.pmid                          ## no prefix
+      ref_doi: relation.article_doi
+      ref_url: relation.bindingdb_link
       ## commenting out because data-processing / biolink-modeling issues
       # in_taxon: subject.organism                 ## was called "Target Source Organism According to Curator or DataSource"
       # bindingdb_curation_datasource: relation.curation_datasource    ## Curation/DataSource
-      # "biolink:source_web_page": relation.bindingdb_link
-      # article_doi: relation.article_doi
       # patent_number: relation.patent_number
     uniprot-subject:
       UniProtKB: subject.uniprot.accession  ## no prefix
       "biolink:original_subject": subject.name
-      pubmed: relation.pmid
+      ref_pmid: relation.pmid                          ## no prefix
+      ref_doi: relation.article_doi
+      ref_url: relation.bindingdb_link
       ## commenting out because data-processing / biolink-modeling issues
       # in_taxon: subject.organism
       # bindingdb_curation_datasource: relation.curation_datasource
-      # "biolink:source_web_page": relation.bindingdb_link
-      # article_doi: relation.article_doi
       # patent_number: relation.patent_number
diff --git a/biolink/openapi.yml b/biolink/openapi.yml
@@ -853,7 +853,7 @@ components:
   x-bte-response-mapping:
     gene-hgnc:
       HGNC: associations.object.HGNC
-      pubmed: associations.publications.id
+      ref_pmid: associations.publications.id
       ## commenting out because data-processing / biolink-modeling issues
       # monarch_source_database: associations.provided_by
       # ## there's also a relation.id and relation.inverse.
@@ -862,68 +862,68 @@ components:
     gene-hgnc-dbsnp-start:
       HGNC: associations.object.HGNC
       input_name: associations.subject.label   ## for the dbsnp input
-      pubmed: associations.publications.id
+      ref_pmid: associations.publications.id
       ## commenting out because data-processing / biolink-modeling issues
       # monarch_source_database: associations.provided_by
       # ## there's also a relation.id and relation.inverse.
       # ## this often has "expressed in" with inverse: true. Meaning the actual relation should be "expresses"
       # "biolink:original_predicate": associations.relation.label
     pathway-react:
       REACT: associations.object.REACT
-      pubmed: associations.publications.id
+      ref_pmid: associations.publications.id
       ## commenting out because data-processing / biolink-modeling issues
       # monarch_source_database: associations.provided_by
       # "biolink:original_predicate": associations.relation.label
     phenotype-hp:
       HP: associations.object.HP
-      pubmed: associations.publications.id
+      ref_pmid: associations.publications.id
       ## commenting out because data-processing / biolink-modeling issues
       # monarch_source_database: associations.provided_by
       # "biolink:original_predicate": associations.relation.label
     phenotype-hp-dbsnp-start:
       HP: associations.object.HP
       input_name: associations.subject.label   ## for the dbsnp input
-      pubmed: associations.publications.id
+      ref_pmid: associations.publications.id
       ## commenting out because data-processing / biolink-modeling issues
       # monarch_source_database: associations.provided_by
       # "biolink:original_predicate": associations.relation.label
     variant-dbsnp:
       DBSNP: associations.object.dbSNP
       output_name: associations.object.label
-      pubmed: associations.publications.id
+      ref_pmid: associations.publications.id
       ## commenting out because data-processing / biolink-modeling issues
       # monarch_source_database: associations.provided_by
       # "biolink:original_predicate": associations.relation.label
     anatomy-uberon:
       UBERON: associations.object.UBERON
-      pubmed: associations.publications.id
+      ref_pmid: associations.publications.id
       ## commenting out because data-processing / biolink-modeling issues
       # monarch_source_database: associations.provided_by
       # "biolink:original_predicate": associations.relation.label
     disease-mondo:
       MONDO: associations.object.MONDO
-      pubmed: associations.publications.id
+      ref_pmid: associations.publications.id
       ## commenting out because data-processing / biolink-modeling issues
       # monarch_source_database: associations.provided_by
       # "biolink:original_predicate": associations.relation.label
     disease-mondo-dbsnp-start:
       MONDO: associations.object.MONDO
       input_name: associations.subject.label   ## for the dbsnp input
-      pubmed: associations.publications.id
+      ref_pmid: associations.publications.id
       ## commenting out because data-processing / biolink-modeling issues
       # monarch_source_database: associations.provided_by
       # "biolink:original_predicate": associations.relation.label
     orthologGene2diseaseMondo:
       MONDO: associations.object.MONDO
-      pubmed: associations.publications.id    ## not all of these are pubmed though...saw some WormBase:WBPaper ones
+      ref_pmid: associations.publications.id    ## not all of these are pubmed though...saw some WormBase:WBPaper ones
       ## commenting out because data-processing / biolink-modeling issues
       # homologous_gene_model_id: associations.subject.id  ## what is directly annotated to the disease
       # homologous_gene_model_name: associations.subject.label
       # monarch_source_database: associations.provided_by
       # "biolink:original_predicate": associations.relation.label
     orthologGene2phenoHP:
       HP: associations.object.HP
-      pubmed: associations.publications.id    ## not all of these are pubmed though...saw some WormBase:WBPaper ones
+      ref_pmid: associations.publications.id    ## not all of these are pubmed though...saw some WormBase:WBPaper ones
       ## commenting out because data-processing / biolink-modeling issues
       # homologous_gene_model_id: associations.subject.id  ## what is directly annotated to the disease
       # homologous_gene_model_name: associations.subject.label

diff --git a/dgidb/openapi.yml b/dgidb/openapi.yml
@@ -1979,14 +1979,14 @@ components:
   x-bte-response-mapping:
     forward:
       NCBIGene: object.NCBIGene
-      pubmed: association.pmids
+      ref_pmid: association.pmids     # no prefix
       ## commenting out because data-processing / biolink-modeling issues
       # dgidb_interaction_claim_source: association.interaction_claim_source
       # "biolink:original_predicate": association.interaction_types  ## or dgidb interaction types
       # dgidb_interaction_group_score: association.interaction_group_score
     reverse:
       "CHEMBL.COMPOUND": subject.CHEMBL_COMPOUND
-      pubmed: association.pmids
+      ref_pmid: association.pmids
       ## commenting out because data-processing / biolink-modeling issues
       # dgidb_interaction_claim_source: association.interaction_claim_source
       # "biolink:original_predicate": association.interaction_types  ## or dgidb interaction types

diff --git a/gtrx/gtrx.yaml b/gtrx/gtrx.yaml
@@ -666,7 +666,7 @@ components:
     drug:
     ## using order and language from gtrx website for these fields
       UNII: object.intervention.inxight   ## no prefix
-      pubmed: references.pmid  ## no prefix
+      ref_pmid: references.pmid  ## no prefix
       ## commenting out because data-processing / biolink-modeling issues
       # type_of_intervention: object.intervention.int_class
       # intervention_MUST_be_started_within: object.timeframe
@@ -679,7 +679,7 @@ components:
       # clinical_summary: subject.clinical_summary
     disease:
       OMIM: subject.omim   ## no prefix
-      pubmed: references.pmid  ## no prefix
+      ref_pmid: references.pmid  ## no prefix
       ## commenting out because data-processing / biolink-modeling issues
       # type_of_intervention: object.intervention.int_class
       # intervention_MUST_be_started_within: object.timeframe

diff --git a/litvar/smartapi.yaml b/litvar/smartapi.yaml
@@ -69,5 +69,4 @@ components:
   x-bte-response-mapping:
     variant_located_in_gene:
       NCBIGene: gene.id  ## no prefix
-      ## commenting out because data-processing / biolink-modeling issues
-      # "biolink:source_web_page": links.url
+      ref_url: links.url