Table of Contents

• About ComptoxAI
  • Authors
  • Citing ComptoxAI
    • Other Publications
  • Source databases
  • Contact Us
  • Contributing
  • Funding and Acknowledgements
• Installing and Building
  • What parts of ComptoxAI should I install?
    • “I just want to explore the data and/or the graph database”
    • “I need to run some Cypher queries on the graph database”
    • “I want to perform machine learning using data from ComptoxAI”
    • “I want to run a few Python analyses using some of ComptoxAI’s data”
    • “I want to create a similar data infrastructure, but for a different purpose”
  • Installing the comptox_ai Python package
  • Installing Neo4j and importing the ComptoxAI graph database
    • Install Neo4j
    • Create an empty graph database
    • Install plugins and configure the database
    • Download the ontology RDF file
    • Import the RDF data into the DBMS
    • Finish tidying up
  • Building and accessing the website
• 8. Adverse Outcome Pathways
  • 8.1. What is an adverse outcome pathway?
  • 8.2. Main components of an AOP
  • 8.3. How are adverse outcome pathways used in ComptoxAI?
  • 8.4. References
• 1. Computational Toxicology
  • 1.1. Methods used in computational toxicology
    • 1.1.1. Quantitative Structure-Activity Relationship (QSAR) modeling
    • 1.1.2. Read-across
    • 1.1.3. Trend Analysis
  • 1.2. Adverse Outcome Pathways
  • 1.3. Major research initiatives in computational toxicology
    • 1.3.1. Tox21
    • 1.3.2. AOP-Wiki / AOP-DB / SAAOP
• 7. Data Schema
  • 7.1. Graph data formats
    • 7.1.1. Graph
    • 7.1.2. Nodes
    • 7.1.3. Relationships
• 6. Source Databases
  • 6.1. Introduction
  • 6.2. References
• 2. Database Queries
  • 2.1. Introduction
  • 2.2. Graph database representation of an entity
  • 2.3. Finding nodes using Cypher
  • 2.4. Finding nodes using Python
  • 2.5. Finding relationships using Cypher
  • 2.6. Finding relationships using Python
• 5. Entity Types in ComptoxAI
  • 5.1. Entity Types
    • 5.1.1. Node types
    • 5.1.2. Relationship types
• 4. Graphs
  • 4.1. Behavior of Neo4j
    • 4.1.1. Induced Subgraphs
• 3. Graph Databases
  • 3.1. Building and installing the database locally
    • 3.1.1. Populating the ontology with individuals
    • 3.1.2. Importing the ontology into Neo4j
  • 3.2. Connecting to our instance of the database
  • 3.3. API Documentation
• User Guide
  • 1. Computational Toxicology
    • 1.1. Methods used in computational toxicology
    • 1.2. Adverse Outcome Pathways
    • 1.3. Major research initiatives in computational toxicology
  • 2. Database Queries
    • 2.1. Introduction
    • 2.2. Graph database representation of an entity
    • 2.3. Finding nodes using Cypher
    • 2.4. Finding nodes using Python
    • 2.5. Finding relationships using Cypher
    • 2.6. Finding relationships using Python
  • 3. Graph Databases
    • 3.1. Building and installing the database locally
    • 3.2. Connecting to our instance of the database
    • 3.3. API Documentation
  • 4. Graphs
    • 4.1. Behavior of Neo4j
  • 5. Entity Types in ComptoxAI
    • 5.1. Entity Types
  • 6. Source Databases
    • 6.1. Introduction
    • 6.2. References
  • 7. Data Schema
    • 7.1. Graph data formats
  • 8. Adverse Outcome Pathways
    • 8.1. What is an adverse outcome pathway?
    • 8.2. Main components of an AOP
    • 8.3. How are adverse outcome pathways used in ComptoxAI?
    • 8.4. References
  • 9. Useful Cypher queries
    • 9.1. Entity search
    • 9.2. Relationship search
    • 9.3. Shortest paths
    • 9.4. Network expansion
    • 9.5. Graph data science algorithms
    • 9.6. Schema-related queries
• Video Tutorials
  • Using the Neo4j database browser
• 9. Useful Cypher queries
  • 9.1. Entity search
  • 9.2. Relationship search
  • 9.3. Shortest paths
  • 9.4. Network expansion
  • 9.5. Graph data science algorithms
  • 9.6. Schema-related queries
• REST Web API
  • API Routes
  • Web API Implementation
  • Contributions and Suggestions
• ComptoxAI Databases
  • GraphDB
    • GraphDB.build_graph_cypher_projection()
    • GraphDB.build_graph_native_projection()
    • GraphDB.convert_ids()
    • GraphDB.drop_all_existing_graphs()
    • GraphDB.drop_existing_graph()
    • GraphDB.export_graph()
    • GraphDB.fetch()
    • GraphDB.fetch_chemical_list()
    • GraphDB.fetch_node_type()
    • GraphDB.fetch_nodes()
    • GraphDB.fetch_relationships()
    • GraphDB.find_node()
    • GraphDB.find_nodes()
    • GraphDB.find_relationships()
    • GraphDB.find_shortest_paths()
    • GraphDB.get_graph_statistics()
    • GraphDB.get_metagraph()
    • GraphDB.list_existing_graphs()
    • GraphDB.run_cypher()
    • GraphDB.stream_named_graph()
  • GraphExporter
    • GraphExporter.GraphDB
    • GraphExporter.stream_subgraph()
  • Node
    • Node.clear()
    • Node.copy()
    • Node.fromkeys()
    • Node.get()
    • Node.items()
    • Node.keys()
    • Node.pop()
    • Node.popitem()
    • Node.setdefault()
    • Node.update()
    • Node.values()
• Graph
  • Graph
    • Graph.add_edges()
    • Graph.add_nodes()
    • Graph.classes()
    • Graph.convert()
    • Graph.edges()
    • Graph.from_dgl()
    • Graph.from_graphsage()
    • Graph.from_neo4j()
    • Graph.from_networkx()
    • Graph.nodes()
  • GraphSAGEData
    • GraphSAGEData.add_edge()
    • GraphSAGEData.add_node()
    • GraphSAGEData.is_heterogeneous
  • NetworkXData
    • NetworkXData.NetworkxJsonEncoder
    • NetworkXData.add_edge()
    • NetworkXData.add_edges()
    • NetworkXData.save_graph()
• Graph Algorithms
• API Reference
  • comptox_ai.db: ComptoxAI’s graph database
    • GraphDB
  • comptox_ai.ml: Machine learning models
    • Graph ML
    • Non-graph ML
• Miscellaneous tools and utilities
• Machine Learning library
• Contact Us