Querying a Neo4j Instance in Python

This instruction set requires having a running Neo4j instance.

Running Neo4j

Neo4j is most easily run locally using the desktop GUI version. Alternatively from the command line it can be run with neo4j start

Importing required modules

from neo4j import GraphDatabase

Initializing Driver and Creating Cypher Query

The driver uses the neo4j-driver python package to connect via neo4j's api to my locally running Reactome Database.

Queries are written the the Cypher query language. A good cheatsheet for commands can be found here

# Initialize connection to database
driver = GraphDatabase.driver('bolt://localhost:7687', auth=('neo4j', 'Akre1234'))

query = 'MATCH (a:Drug) RETURN a.name, a.stId LIMIT 5'

Run Cypher query and print desired results

The query returned name and stId properties of 5 drugs. Each returned record will only contain the specified information form the query.

with driver.session() as session:
    info = session.run(query)
    for item in info:
        print('Name:', item.values()[0], 'id:', item.values()[1])

Name: ['trastuzumab', 'herceptin', 'D5v8', 'R-597'] id: R-ALL-9634466
Name: ['CP-724714'] id: R-ALL-9649889
Name: ['Afatinib', 'BIBW2992', 'Irreversible TKI inhibitor afatinib generic inhibits  EGFR and  ERBB2 (HER2)'] id: R-ALL-1220577
Name: ['AZ5104'] id: R-ALL-9649879
Name: ['Sapitinib'] id: R-ALL-9649894