Directed Small Molecules Network (DSMN)

Working with the DSMN data in Neo4j:

** Note: This documentation was written for Neo4j version: community-3.5.7 (which requires Java 8 to run!).

Find our tutorial website here. Download the community edition of Neo4j (for free) here. The download on Windows machines is quite straightforward, for Linux see here.

1. Run Neo4j

1.1 Windows:

• Most likely, your computer will create a shortcut on your desktop.
• Double click the icon, ans press the start button (if you cannot start Neo4j, check your version of Java!).

1.2 Linux

(read the README.txt for updated instructions, when using a different version): - On your computer, locate the location where you have downloaded Neo4j. - Open up your terminal, and navigate to this location. - Type "./bin/neo4j console" and hit enter to start (if you cannot start Neo4j, check your version of Java!). - Shutdown the server by typing Ctrl-C in the console.

2. Download DSMN data

Find a copy of the graph.db file here and unzip the file.

3. Load pre-existing database:

3.1 Windows

(Note: you can also use the command line for the steps listed below; see section '3.2 Linux' for more detailed instructions). - Start up the Neo4j instance - For first time users: you can see a user interface of Neo4j in your internetbrowser by adding this location in the search bar: http://localhost:7474/ (see section '4. Visualise and interact with data' for more details). Login the first time with the username 'neo4j' and password 'neo4j'; after this you will be prompted to submit a personal password. For next usages (also in Cytoscape with the CyNeo4j app), you will use the username 'neo4j', and the personal password you picked. - Find the location on your computer where you downloaded the graph.db file. - Select the correct database location folder, and click on “start” - After a few seconds the status bar will turn from red to green and displays the message: “Neo4j is ready…”

3.2 Linux

For first time users, first start your neo4j instance: shell cd PATH/TO/NEO4J ./bin/neo4j console

Then, view the user interface of Neo4j in your internetbrowser by adding this location in the search bar: http://localhost:7474/ . Login the first time with the username 'neo4j' and password 'neo4j'; after this you will be prompted to submit a personal password. For next usages (also in Cytoscape with the CyNeo4j app), you will use the username 'neo4j', and the personal password you picked.

After the initial setup (setting your personal password), you can follow the following instructions: shell cd PATH/TO/NEO4J/data/databases rm -r graph.db cp -r /PATH/TO/DOWNLOADED_DATA/DSMN_MONTHYEAR.graph.db/graph.db/ /PATH/TO/NEO4J/data/databases/ cd ../.. ./bin/neo4j console Open the remote interface (hhtp://localhost:7474) and login with your personal password.

4. Visualise and interact with data

Windows+Linux: - To see your Neo4j data, open http://localhost:7474/ in an (internet)browser. - To see which data is loaded, click on the database icon (Top left corner, green in image below). - To create a simple query, click on one of the Node/Interaction Labels or property types. The Neo4j server automatically creates a query and displays the results, see image below for "Metabolites".

Working with the CyNeo4j App:

5. Download the CyNeo4j app

From the Cytoscape App store here or from within Cytoscape.

6. Connect to the Neo4j database

Note: you need to have a Neo4j instance running.

Under 'Apps/cyNeo4j', select the option called 'Connect to instance'. Fill in your username and personal password, and click 'Connect'.

7. Run your queries

Note: accepted IDs for metabolites are Wikidata, ChEBI, and HMDB.

1. Under 'Apps/cyNeo4j', select the option called 'Directed Small Molecules Network'. Fill in the IDs that you want to retrieve a subnetwork for (examples can be found here), one ID per line, and click 'Submit'. The shortest path calculation will now start, and the results visualized in individual Cytoscape networks (note that this might take a few seconds, depending on the size of your input query).
2. Use the button located right under the main menu bar Fill in the IDs that you want to retrieve a subnetwork for (examples can be found here), one ID per line, and click 'Submit'. The shortest path calculation will now start, and the results visualized in individual Cytoscape networks (note that this might take a few seconds, depending on the size of your input query).

8. Results interpretation

The lefthand side menu (1) shows you which networks you have build, the middle top panel (2) shows the network from the shortest path query itself, the right panel (3) and overview of the results for the shortest path query, and the middle bottom panel (4) a table overview of the retrieved information (notice that you can switch here between Node, Edge, and Network Table).

If you wish to create the DSMN data in a Neo4j Graph database yourself, follow the following steps:

GitHub repository to work on the Directed Small Molecules Network project.

Minimum System Requirements: - Java 8 (Eclipse, Neo4j) - WikiPathways RDF Nov. 2020 GPML2RDF

Compiling the script

The setup of this project in Eclipse has been tested with: * Eclipse Java Neon, OS Windows 7 * Eclipse IDE 2019-03, OS Linux (Debian)

From the command line

The code can also be compiled from the command line:

shell mvn clean install