PretermMetaproteomeGraphDB

Dockerized graph database as referenced in:

J. Alfredo Blakeley-Ruiz, Stephen K. Grady, Carissa Bleker, Michael A. Langston, and Robert L. Hettich (2020) Highly connected metabolic networks and discriminatory reactions for necrotizing enterocolitis extracted from metaproteomics of preterm human infant gut microbiomes. (forthcoming)

Feel free to create an issue with questions.

Data

Minimal data is kept in this repo. Files are avalible in the Massive repository, identifier MSV000086096.

Set up

Part 1: Neo4j Database

To create a copy of the graphDB on you own computer you will need docker and docker-compose.

1. Clone this repo to your computer.

   git clone https://github.com/UTKLangstonLab/PretermMetaproteomeGraphDB.git 
   

2. Build the neo4j container with the graph. (It may take a few minutes to download the graph dump. )

   cd PretermMetaproteomeGraphDB
   docker build --tag proteomegraph .
   

3. Run the image:

   docker run -it  -p 7474:7474 -p 1337:1337 -p 7687:7687 proteomegraph
   

And visit http://localhost:7474/.

Part 2: Link the neo4j database to a Jupyter notebook

Start up the neo4j database and linked jupyter notebook with docker-compose. To run the first time, navigate to the folder (containing docker-compose.yml) and type:

docker-compose up

Thereafter you can use

docker-compose start

To stop the containers, use:

docker-compose stop

Or to stop and remove them:

docker-compose down

Open your browser at the http://localhost:8888/?token=... link printed to the terminal. If the logs are not printed, run

docker-compose logs | grep 'http://localhost:.*/?token' | tail -1 

to find the correct link.

To remove

If you need to remove the graph image.

1. Remove the container:

   docker-compose down
   

   Or get the container ID from the first column in:

   docker ps -a | grep proteomegraph
   

   And delete it using:

   docker rm <container ID>  
   

2. Delete the image:

   docker rmi proteomegraph
   

3. Delete the folder with the graph data (if it exists).

   sudo rm -rf ~/neo4j/data/proteome/