Improve DKG assembly of physical constants

mira/dkg/askemo/askemo.climate.json

@@ -107,6 +107,24 @@
     ],
     "type": "class"
   },
+  {
+    "description": "",
+    "id": "askem.climate:0000202",
+    "name": "Lie derivative",
+    "parents": [
+      "askem.climate:0000009"
+    ],
+    "part_ofs": [
+      "askem.climate:0001006"
+    ],
+    "synonyms": [
+      {
+        "type": "referenced_by_symbol",
+        "value": "ℒ"
+      }
+    ],
+    "type": "class"
+  },
   {
     "description": "",
     "id": "askem.climate:0001001",
@@ -144,12 +162,33 @@
     "description": "",
     "id": "askem.climate:0001004",
     "name": "Momentum conservation equation",
+    "parents": [
+      "askem.climate:0000006"
+    ],
     "type": "class"
   },
   {
     "description": "",
     "id": "askem.climate:0001005",
     "name": "tracer conservation",
+    "parents": [
+      "askem.climate:0000006"
+    ],
+    "type": "class"
+  },
+  {
+    "description": "Vegetation dynamics in semi-arid environments",
+    "id": "askem.climate:0001006",
+    "name": "Klausmeier",
+    "type": "class"
+  },
+  {
+    "description": "Composition of Halfar + Glen's law",
+    "id": "askem.climate:0001007",
+    "name": "Ice Dynamics",
+    "parents": [
+      "askem.climate:0000006"
+    ],
     "type": "class"
   },
   {
@@ -283,6 +322,26 @@
     ],
     "type": "class"
   },
+  {
+    "description": "Water growth and dispursion in semi-arid environments",
+    "dimensionality": "kg m^-2",
+    "id": "askem.climate:0002010",
+    "name": "hydrodynamics",
+    "part_ofs": [
+      "askem.climate:0001006"
+    ],
+    "type": "class"
+  },
+  {
+    "description": "Plant growth and dynamics in semi-arid environments",
+    "dimensionality": "kg m^-2",
+    "id": "askem.climate:0002011",
+    "name": "phytodynamics",
+    "part_ofs": [
+      "askem.climate:0001006"
+    ],
+    "type": "class"
+  },
   {
     "description": "A coordinate representing the amount north or south of the equator on Earth",
     "dimensionality": "radians",
@@ -1022,5 +1081,117 @@
       }
     ],
     "type": "class"
+  },
+  {
+    "description": "How much water is there in some area",
+    "dimensionality": "kg m^-2",
+    "id": "askem.climate:0003046",
+    "name": "water density",
+    "part_ofs": [
+      "askem.climate:0002010"
+    ],
+    "synonyms": [
+      {
+        "type": "referenced_by_symbol",
+        "value": "w"
+      }
+    ],
+    "type": "class"
+  },
+  {
+    "description": "Water input to the model",
+    "dimensionality": "kg m^-2",
+    "id": "askem.climate:0003047",
+    "name": "water input",
+    "part_ofs": [
+      "askem.climate:0002010"
+    ],
+    "synonyms": [
+      {
+        "type": "referenced_by_symbol",
+        "value": "a"
+      }
+    ],
+    "type": "class"
+  },
+  {
+    "description": "Also called water flow speed",
+    "dimensionality": "m s^-1",
+    "id": "askem.climate:0003048",
+    "name": "rate of water flow downhill",
+    "part_ofs": [
+      "askem.climate:0002010"
+    ],
+    "synonyms": [
+      {
+        "type": "referenced_by_symbol",
+        "value": "ν"
+      }
+    ],
+    "type": "class"
+  },
+  {
+    "description": "",
+    "id": "askem.climate:0003049",
+    "name": "change in water density in downhill direction",
+    "part_ofs": [
+      "askem.climate:0002010"
+    ],
+    "synonyms": [
+      {
+        "type": "referenced_by_symbol",
+        "value": "ℒ(dX, w)"
+      }
+    ],
+    "type": "class"
+  },
+  {
+    "description": "How much vegetable mass there is in some area",
+    "dimensionality": "kg m^-2",
+    "id": "askem.climate:0003050",
+    "name": "vegetation density",
+    "part_ofs": [
+      "askem.climate:0002011",
+      "askem.climate:0002010"
+    ],
+    "synonyms": [
+      {
+        "type": "referenced_by_symbol",
+        "value": "n"
+      }
+    ],
+    "type": "class"
+  },
+  {
+    "description": "Contributors to plant death, linear in plant density",
+    "dimensionality": "unitless",
+    "id": "askem.climate:0003051",
+    "name": "plant loss coefficient",
+    "part_ofs": [
+      "askem.climate:0002011"
+    ],
+    "synonyms": [
+      {
+        "type": "referenced_by_symbol",
+        "value": "m"
+      }
+    ],
+    "type": "class"
+  },
+  {
+    "description": "",
+    "dimensionality": "kg m^-2",
+    "id": "askem.climate:0003052",
+    "name": "diffusion of vegetation",
+    "part_ofs": [
+      "askem.climate:0002011"
+    ],
+    "synonyms": [
+      {
+        "type": "referenced_by_symbol",
+        "value": "Δ(n)"
+      }
+    ],
+    "type": "class"
   }
 ]

mira/dkg/askemo/askemo_climate.py

@@ -21,8 +21,7 @@
 
 def get_askem_climate_terms() -> Dict[str, Term]:
     """Get ASKEM Climate ontology terms."""
-    # df = pd.read_csv(URL, sep="\t")
-    df = pd.read_excel("/Users/cthoyt/Downloads/ASKEM Climate Ontology.xlsx")
+    df = pd.read_csv(URL, sep="\t")
     df = df[df["curie"].notna()]
     df.columns = [c.lower() for c in df.columns]
 

mira/dkg/construct.py

@@ -329,7 +329,7 @@ def construct(
                     xref.type or "oboinowl:hasDbXref" for xref in term.xrefs or []
                 ),
                 synonym_types=";".join(
-                    synonym.type or "skos:exactMatch" for synonym in term.synonyms or []
+                    synonym.type or "oboInOwl:hasExactSynonym" for synonym in term.synonyms or []
                 ),
             )
             for parent_curie in term.parents:
@@ -548,28 +548,44 @@ def construct(
         )
 
     click.secho("Physical Constants", fg="green", bold=True)
-    for wikidata_id, label, description, synonyms, xrefs, _value, _formula, _symbols in tqdm(
+    for wikidata_id, label, description, synonyms, xrefs, value, formula, symbols in tqdm(
         get_physical_constant_terms(), desc="Physical Constants"
     ):
-        # TODO how to model value, formula?
-        # TODO process mathml and make readable
         curie = f"wikidata:{wikidata_id}"
         node_sources[curie].add("wikidata")
+
+        prop_predicates, prop_values = [], []
+        if value:
+            prop_predicates.append("debio:0000042")
+            prop_values.append(str(value))
+        # TODO process mathml and make readable
+        # if formula:
+        #     prop_predicates.append("debio:0000043")
+        #     prop_values.append(str(formula))
+
+        synonym_types, synonym_values = [], []
+        for syn in synonyms:
+            synonym_values.append(syn)
+            synonym_types.append("oboInOwl:hasExactSynonym")
+        for symbol in symbols:
+            synonym_values.append(symbol)
+            synonym_types.append("debio:0000031")
+
         nodes[curie] = NodeInfo(
             curie=curie,
             prefix="wikidata;constant",
             label=label,
-            synonyms=";".join(synonyms),
+            synonyms=";".join(synonym_values),
+            synonym_types=";".join(synonym_types),
             deprecated="false",
             type="class",
             definition=description,
             xrefs=";".join(xrefs),
             xref_types=";".join("oboinowl:hasDbXref" for _ in xrefs),
             alts="",
             version="",
-            property_predicates="",
-            property_values="",
-            synonym_types="",
+            property_predicates=";".join(prop_predicates),
+            property_values=";".join(prop_values),
         )
 
     def _get_edge_name(curie_: str, strict: bool = False) -> str:
@@ -1050,7 +1066,7 @@ def get_node_info(term: pyobo.Term, type: EntityType = "class"):
         property_values="",
         xref_types="",
         synonym_types=";".join(
-            synonym.type.curie if synonym.type is not None else "skos:exactMatch"
+            synonym.type.curie if synonym.type is not None else "oboInOwl:hasExactSynonym"
             for synonym in term.synonyms or []
         ),
     )

mira/dkg/resources/cso.py

@@ -1,3 +1,4 @@
+from pyobo import Obo
 from pyobo.reader import from_obo_path
 from pystow.utils import download
 import bioregistry
@@ -12,7 +13,7 @@
 URL = "https://raw.githubusercontent.com/cthoyt/Climate-System-Ontology-CSO/main/cso.obo"
 
 
-def get_cso_obo():
+def get_cso_obo() -> Obo:
     bioregistry.manager.synonyms["cso"] = "cso"
     bioregistry.manager.registry["cso"] = bioregistry.Resource(
         name="Climate Systems Ontology",
@@ -21,6 +22,7 @@ def get_cso_obo():
         description="An ontology for climate systems",
     )
     download(url=URL, path=PATH)
+    # use https://github.com/pyobo/pyobo/pull/159
     return from_obo_path(PATH, prefix="cso", default_prefix="cso", strict=False)