NCATS-Tangerine/stringdb-beacon

Name: stringdb-beacon

Owner: NCATS Data Translator Project - Tangerine Team

Description: Knowledge Beacon Wrapper for the StringDb resource (https://string-db.org/)

Created: 2017-06-26 20:10:37.0

Updated: 2017-09-21 21:46:27.0

Pushed: 2017-11-09 04:29:45.0

Homepage: null

Size: 860

Language: Python

GitHub Committers

User	Most Recent Commit	# Commits

Other Committers

User	Email	Most Recent Commit	# Commits

README

stringdb-beacon

Knowledge Beacon Wrapper for the StringDb resource (https://string-db.org/)

Overview

This server was generated by the swagger-codegen project. By using the OpenAPI-Spec from a remote server, you can easily generate a server stub. This is an example of building a swagger-enabled Flask server.

This example uses the Connexion library on top of Flask.

Running with Docker

Use Docker Compose to build and run the application. Before using compose in the project, you should first copy the configuration file:

ocker-compose.yaml-template docker-compose.yaml

The host machines NEO4J_AUTH environment variable is passed into the docker containers. If it is not set then it will default to neo4j/password. You can also directly modify the default password (plus Docker port redirections, etc.) in this 'docker-compose.yaml' file.

Once ready to build, type the following into the terminal::

rt NEO4J_AUTH=neo4j/<password>  # if you wish to change your default password
tringdb-beacon
er-compose build
er-compose up

Note: If you don't wish to set a password, you can simply set NEO4J_AUTH=none. You can control which ports, if any at all, are exposed with the docker-compose.yaml file.

The Neo4j container's data, import, logs, and config directories will be mounted on the host machine at $HOME/neo4j. All the data in the database, including the username and password, will persist in the $HOME/neo4j/data directory over multiple runs of the container.

Note: If the docker-compose commands are is giving you trouble, try running them as the system administrator with the sudo command. Remember, though, that if you are running your commands as 'sudo', then depending on how your Linux instance configures sudo, in some cases, $HOME may actually be '/root' or it may otherwise still be in your normal user home. This may consequences for data importation (see below)

The REST API should now be running at http://localhost:5000/api, the Swagger web interface at http://localhost:5000/api/ui, and the Neo4j browser user interface at http://localhost:7474. You can open your browser with these addresses to see these applications in action. You wont see much until you load data into the database, though.

Loading Data

String-db hosts its database at https://string-db.org/cgi/download.pl. You can get subsets of the database by selecting a species, so far we have only been using Homo sapiens. There is a python script neo4j/download_stringdb_data.py that can be used to download, unzip, and properly format the data for Homo sapiens. This script must be run with version 3 of python, in the terminal run python --version to check your version. If you will not be using this script you will need to manually fix TSV headers in some of the data files.

tringdb-beacon/neo4j
on3 download_stringdb_data.py

If you have used docker-compose to run the application, you will have a neo4j/import directory mounted in your $HOME directory. If you are running docker-compose as 'sudo', this directory may be '/root' in which case, you should may also run the download script as 'sudo' so thta $HOME defaults to '/root', which is where your Docker neo4j database will expect to see the data. If your $HOME directory is still your normal user home directory, you can likely still run this download script without sudo.

Your Neo4j docker container should be named stringdbbeacon_db_1, run docker-compose ps to confirm this. We will execute the load.cql file, which was copied over to the 'import' directory with the downloaded data, sing the Neo4j shell.

Before loading the data, you first turn off the docker container (from within the stringdb-beacon directory):

 docker-compose stop db

This will allow data loading directly using the neo4j-shell script, as follows:

 docker run --env=NEO4J_dbms.directories.import=/import --env=NEO4J_dbms_memory_heap_maxSize=2G  --volume=$HOME/neo4j/import:/import --volume=$HOME/neo4j/data:/data neo4j:latest /var/lib/neo4j/bin/neo4j-shell -path /data/databases/graph.db -file /import/load.cql

Note: Again, if running other commands as 'sudo', then run this command as 'sudo' as well. If you see java.lang.OutOfMemoryError: Java heap space, then you may need to increase the Java heap size with the NEO4J_dbms_memory_heap_maxSize environment variable noted above.

After loading, the database docker container may be restarted:

 docker-compose up db

and the data should now be visible through the Neo4j web client on port 7474.

Issues

This beacon is a work in progress. Here are a few points that still need to be addressed

• So far the evidence and exactmatches API are not implemented.
• I have only been using data for Homo sapiens. We should also include mouse, rat, yeast, c. elegans, drosphila, and zebrafish species as well, to complement monarch biolink.
• Only a very small subset of the database files are being used.