neo4j operations

schnellerereETL.py

@@ -2,6 +2,49 @@ import json
 import pandas as pd
 from pathlib import Path
 from collections import defaultdict
+from neo4j import GraphDatabase
+
+NEO4J_URI = "bolt://localhost:7687"
+NEO4J_USER = "neo4j"
+NEO4J_PASSWORD = "password"
+
+driver = GraphDatabase.driver(
+    NEO4J_URI,
+    auth=(NEO4J_USER, NEO4J_PASSWORD)
+)
+
+
+
+def load_nodes(df, label):
+    with driver.session() as session:
+        for _, row in df.iterrows():
+            session.run(
+                f"""
+                MERGE (n:{label} {{id: $id}})
+                SET n += $props
+                """,
+                id=row["id"],
+                props=row.drop("id").dropna().to_dict()
+            )
+
+
+def load_edges(edges_df):
+    with driver.session() as session:
+        for _, row in edges_df.iterrows():
+            session.run(
+                """
+                MATCH (s {id: $source})
+                MATCH (t {id: $target})
+                CALL apoc.create.relationship(s, $type, {}, t)
+                YIELD rel
+                RETURN rel
+                """,
+                source=row["source"],
+                target=row["target"],
+                type=row["type"].upper()
+            )
+
+
 
 
 # KONFIGURATION
@@ -433,6 +476,19 @@ network_edges_df = pd.DataFrame(network_edges)
 network_nodes_df.to_csv(OUTPUT_DIR / "network_nodes.csv", index=False)
 network_edges_df.to_csv(OUTPUT_DIR / "network_edges.csv", index=False)
 
+
+
+load_nodes(nodes_df[nodes_df['kind']=="Gene"], "Gene")
+load_nodes(nodes_df[nodes_df['kind']=="Disease"], "Disease")
+load_nodes(nodes_df[nodes_df['kind']=="Compound"], "Compound")
+load_nodes(nodes_df[nodes_df['kind']=="Symptom"], "Symptom")
+load_nodes(nodes_df[nodes_df['kind']=="Side Effect"], "SideEffect")
+
+
+load_edges(edges_df)
+
+
+
 print(f"Network: {len(network_nodes_df)} nodes, {len(network_edges_df)} edges")