Omnipath Secondary Adapter

This repository contains the scripts to run OntoWeaver and BioCypher for the creation of knowledge graphs from the OmniPath database. The project is currently under development.

Overview

The repository contains an Ontoweaver mapping file that serves as a BioCypher "secondary adapter" for OmniPath database. The adapter includes predefined sets of node types (entitites) and edge types (relationships) for the different tables of Omnipath database (Networks, Enzyme-PTM, Complexes, Annotations, Intercell). A script is also provided to run OntoWeaver and BioCypher to create the knowledge graph.

About OmniPath

Omnipath is a database of molecular biology prior knowledge developed in Saez Lab and Korcsmaros Lab. It combines data from more than 100 resources and contains:

• protein-protein and gene regulatory interactions
• Enzyme and Post-Translational-Modifications(PTM) relationships
• Protein complexes
• Protein annotations
• intercellular communication.

Omnipath database stores the information in five different tables containing the aforementioned data.

• Networks
• Enzyme-PTM
• Complexes
• Annotations
• Intercell

#> [!WARNING] #> This BioCypher adapter relies on the Ontoweaver library that works with Python 3.12. Ensure the #Python version is 3.12 in your virtual environment.

About this adapter

The main goal of this adapter is to enable users to retrieve data from the Omnipath database and use that information to generate a knowledge graph. For achieving this goal, we are going use BioCypher and OntoWeaver.

• BioCypher:it is a framework for building biomedical knowledge graphs by integrating heterogeneous data into a structured, ontology-aligned graph representation.

• OntoWeaver: it is a visual and declarative ontology-based tool for designing and managing data integration systems, enabling intuitive schema mapping and semantic consistency. OntoWeaver has been built built on top of Biocypher.

Prerequisites

• Python 3
• Poetry [recommended]: Python packaging and dependency manager.
  • Install Poetry
• git: version control manager
  • Install git

Prerequisite	Version	Verify installation	How to install?
Python 3	>=3.12	$ python --version	link
Poetry	1.8	$ poetry about	link
git	>= 2.0	$ git --version	link
Neo4j server	2025.04.0	$ neo4j --version	link

Usage

1. Clone this repository and change to the directory:

git clone https://github.com/biocypher/omnipath-secondary-adapter
cd omnipath-secondary-adapter

2. Install all the dependencies (pre-configured in the pyproject.toml file):

poetry lock
poetry install --no-root

3. This current implementation generates Neo4j scripts for each table in the Omnipath database

Omnipath table	Script parameter	Neo4j script generation
Networks	-net or --networks	✅ Done
Enzyme-PTM	-enz or --enzyme-PTM	✅ Done
Complexes	-co or --complexes	✅ Done
Annotations	-an or --annotations	✅ Done
Intercell	-inter or --intercell	✅ Done

Use the following commands to generate import scripts for each Omnipath table:

Networks

poetry run python weave_knowledge_graph.py -net download

Enzyme-PTM

poetry run python weave_knowledge_graph.py -enz download

Complexess

poetry run python weave_knowledge_graph.py -co download

Annotations

poetry run python weave_knowledge_graph.py -an download

Intercell

poetry run python weave_knowledge_graph.py -inter download

4. Once the script has processed the data, you can verify a folder has been generated under biocypher-out. The folder is named after the created date (e.g., 20250610172616). This folder contains the following:

• neo4j-admin-import-call.sh: script to import the the data in Neo4j.
• CSV files:
  • *-header.csv: indicate the node/edge properties fields
  • *-part000.csv, *-part001.csv,...: store nodes/edges data based on the structure indicated in the *-header.csv files.

5. Stop the running instances of Neo4j by running:

sudo neo4j stop

6. Run the import script generated by step 4.

sudo bash biocypher-out/<datetime>/neo4j-admin-import-call.sh

7. Start again Neo4j by running:

sudo neo4j start

8. Open the Neo4j platform on http://localhost:7474. You will be ask for credentials. By default the Neo4j credentials are:

• Database User: neo4j
• Password: neo4j

Note: change the password as requested.

9. At the end, you have you Knowledge Graph! 🎉 Congratulations!