Add ncbitaxon:has_division property.

src/ncbitaxon.py

@@ -154,8 +154,9 @@ def convert_synonyms(tax_id, synonyms):
     return output
 
 
-def convert_node(node, label, merged, synonyms, citations):
+def convert_node(node, label, merged, synonyms, citations, divisions):
     """Given a node dictionary, a label string, and lists for merged, synonyms, and citations,
+    and a divisions dictionary mapping division IDs to names,
     return a Turtle string representing this tax_id."""
     tax_id = node["tax_id"]
     output = [f"NCBITaxon:{tax_id} a owl:Class"]
@@ -193,6 +194,13 @@ def convert_node(node, label, merged, synonyms, citations):
     gc_id = node["genetic_code_id"]
     if gc_id:
         output.append(f'; oboInOwl:hasDbXref "GC_ID:{gc_id}"^^xsd:string')
+
+    div_id = node["division_id"]
+    if div_id and div_id in divisions:
+        division_name= escape_literal(divisions[div_id])
+        output.append(f'; ncbitaxon:has_division "{division_name}"^^xsd:string')
+        # division_id = label_to_id(divisions[div_id])
+        # output.append(f'; ncbitaxon:has_division NCBITaxon:{division_id}')
 
     for merge in merged:
         output.append(f'; oboInOwl:hasAlternativeId "NCBITaxon:{merge}"^^xsd:string')
@@ -224,6 +232,7 @@ def convert(taxdmp_path, output_path, taxa=None):
     synonyms = defaultdict(list)
     merged = defaultdict(list)
     citations = defaultdict(list)
+    divisions = defaultdict(str)
     with open(output_path, "w") as output:
         isodate = date.today().isoformat()
         ncbi_date = date.today().replace(day=1)
@@ -299,7 +308,18 @@ def convert(taxdmp_path, output_path, taxa=None):
                         rdfs:subPropertyOf oboInOwl:SynonymTypeProperty .
                 """))
 
+        output.write("""ncbitaxon:has_division a owl:AnnotationProperty
+; rdfs:label "has division"^^xsd:string
+; rdfs:comment "A metadata relation indicating taxonomic division (eg Bacteria, Eukaryota)"^^xsd:string
+; oboInOwl:hasOBONamespace "ncbi_taxonomy"^^xsd:string
+.
+""")
         with zipfile.ZipFile(taxdmp_path) as taxdmp:
+            with taxdmp.open("division.dmp") as dmp:
+                for line in io.TextIOWrapper(dmp):
+                    div_id, _div_code, name, _comments , _ = split_line(line)
+                    divisions[div_id] = name
+
             with taxdmp.open("names.dmp") as dmp:
                 for line in io.TextIOWrapper(dmp):
                     tax_id, name, unique, name_class, _ = split_line(line)
@@ -365,6 +385,7 @@ def convert(taxdmp_path, output_path, taxa=None):
                         merged[tax_id],
                         synonyms[tax_id],
                         citations[tax_id],
+                        divisions,
                     )
                     output.write(result)
 
@@ -424,6 +445,17 @@ def convert(taxdmp_path, output_path, taxa=None):
 ; obo:IAO_0100001 {rank_curie}
 .
 """
+#             )
+#         # Add division classes
+#         for division_id, division_name in divisions.items():
+#             division_class_id = label_to_id(division_name)
+#             output.write(
+#                 f"""NCBITaxon:{division_class_id} a owl:Class
+# ; rdfs:label "{division_name}"^^xsd:string
+# ; rdfs:subClassOf <http://purl.obolibrary.org/obo/NCBITaxon#_taxonomic_division>
+# ; oboInOwl:hasOBONamespace "ncbi_taxonomy"^^xsd:string
+# .
+# """
             )
 
             parent_taxrank_id = "9000000" if label in pseudo_ranks else "0000000"
@@ -460,3 +492,12 @@ def main():
 
 if __name__ == "__main__":
     main()
+
+
+# commented chunk lines 368 to 373
+# ; oboInOwl:hasOBONamespace "ncbi_taxonomy"^^xsd:string
+# .
+
+# <http://purl.obolibrary.org/obo/NCBITaxon#_taxonomic_division> a owl:Class
+# ; rdfs:label "taxonomic division"^^xsd:string
+# ; rdfs:comment "This is an abstract class for NCBI taxonomic divisions."^^xsd:string