Repository containing or linking to all resources necessary to build a fully featured MaSyMoS Docker Image
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
owl
patches
scripts
src
.gitattributes
.gitignore
.gitmodules
Dockerfile
Dockerfile.build-env
README.md
build.sh

README.md

MaSyMoS Docker Image

This git repository contains all scripts and links to resources for building a fully featured MaSyMoS Docker image. MaSyMoS is a graph database for SystemsBiology models, developed within the SEMS Project (Repoistories).

This image includes following MaSyMoS modules:

Use the Image

Using the image relatively straight forward. A precompiled version is available on Docker Hub. Simply run:

docker run -it -p 7474:7474 -p 7687:7687 -v /path/to/masymos-data:/data freakybytes/masymos:latest

On the first startup the MaSyMoS image will import several ontologies into the underlaying neo4j database. These are relevant for interlinking models and build better indexes. However this will take quite some time. (~2-4h)

Configure the Image

Since this images is based on the official Neo4j Docker image all configuration mechanisms are also applicable to this image. But keep in mind, that eg. mounting /config will override the MaSyMoS specific Neo4j config. For an overview of which configuration parameter were changed, refer to the patches/ directory.

Build the Image

Due to the extensive build process and the related consumtion of disk space for caching, we decided to split the process into two images: The first one contains a build environment (Dockerfile.build-env) and the actual container running MaSyMoS (Dockerfiler). Unfortunetly this implies, that the image is unsuitable for automated builds on Docker Hub. However to simplify building the image, we're shipping a bash script automating the steps.

Before cloning this repository please ensure, that you have Git LFS installed, since we use it to store the ontology files in this repository. Then clone the repository with all sub-modules:

git clone --recursive git@github.com:SemsProject/masymos-docker.git

The easy way

The simplest way to reproducibly build this MaSyMoS image is to run the shipped build script.

./build.sh

The slightly more difficult way

Alternatively you may run the steps from build.sh yourself.

Start by building an image containing the build environment for MaSyMoS:

docker build -f Dockerfile.build-env -t masymos-build-env:latest .

Next let's utilize this environment to actually compile all MaSyMoS modules. Doing so will run scripts/build_masymos.sh within the container. However to actually persist the build result on your hard drive, we need to mount the src directory as volumne.

docker run --rm -t -v $(pwd)/src/:/root/src/ -e "SRC_DIR=/root/src"  masymos-build-env:latest

Finally we can build the actual MaSyMoS image

docker build -f Dockerfile -t freakybytes/masymos:latest .

Congratulations you're now able to use the image!