Allowing for more data incompleteness

lotusnprod · Oct 31, 2023 · bc43efd · bc43efd
1 parent ce9c76e
commit bc43efd
Show file tree

Hide file tree

Showing 2 changed files with 11 additions and 2 deletions.
diff --git a/update/download_couples_referenced.py b/update/download_couples_referenced.py
@@ -8,7 +8,8 @@ def run(root: Path, retry: int = 3) -> None:
     query = """
     SELECT DISTINCT ?compound ?taxon ?reference WHERE {
       ?compound p:P703 ?statement;
-        wdt:P233 ?canonical_smiles.
+        # We use InChIKey (P235) instead of SMILES as some of them are incomplete.
+        wdt:P235 [].
       ?statement ps:P703 ?taxon;
         (prov:wasDerivedFrom/pr:P248) ?reference.
     }

diff --git a/update/download_smiles.py b/update/download_smiles.py
@@ -8,7 +8,15 @@ def run(root: Path, retry: int = 3) -> None:
     query = """
     SELECT DISTINCT ?structure ?structure_smiles ?canonical_smiles WHERE {
         ?structure wdt:P703 ?taxon;
-                   wdt:P233 ?canonical_smiles.
+          # Using InChIKey (P235) to recognize chemicals. 
+          # Could also be
+          # P31 wd:Q113145171 `type of a chemical entity`
+          # P31 wd:Q59199015 `group of stereoisomers`
+          wdt:P235 [].
+        # All P2017 should also have P233 but some of them are not complete.
+        OPTIONAL {
+            ?structure wdt:P233 ?canonical_smiles.
+        }           
         OPTIONAL {
             ?structure wdt:P2017 ?structure_smiles.
         }