This project provides a framework for large scale phylogenetic analysis in the form of a Web API, a graph oriented database (Neo4j), and a plugin for the latter. The goal of this project is to allow the representation of large phylogenetic networks and trees, as well as ancillary data, support queries on such data, and the deployment of algorithms for inferring/detecting patterns and for computing visualizations. It started being developed in the scope of a master thesis at IST (Instituto Superior Técnico). The unit tests and benchmarks developed are available in the test folder.
The following software tools must be available in your system as they will be called by our scripts.
- Java Development Kit 17 or greater (
javaandjavac) - Maven (
mvn) - Gradle (
gradle) - Docker (
docker,docker-composeand preferably thedocker buildxcommand)
Configuration scripts will check if these are available.
For Linux/UNIX systems, please ensure that a docker user group exists in your system and that the executing user belongs to it.
git clone https://github.com/phyloviz/phyloDB.git phyloDB.git
Note that you might have to adjust the file application.properties, in particular the Neo4j bolt connection.
cd phyloDB.git
./configure.sh # build phyloDB's JAR files.
cd docker
./build-docker.sh # fetch latest stable Neo4j APOC libraries and build the phyloDB Docker image.
At this point, if no issue has occurred, the build process will be at step four of the Wiki guide. The Wiki of this project documents several topics, namely archictecural views, deployment, authentication and API definition (usage). As previously mentioned, the deployment steps present in the wiki can be skiped to step 4 if you have done the previous steps that automatize steps 1 to 3.
After performing the build steps, the phylodb Docker containers may be launched (you should be in directory phyloDB.git/docker):
./start-docker.sh
This will launch the following containers:
- phylodb-app: contains the phyloDB server logic.
- phylodb-neo4j: contains the Neo4j database where project data is stored.
The Docker container internal files will be mapped to the directory phyloDB.git/docker/instance in your local machine:
phyloDB.git/docker/instance/app: files corresponding to the phyloDB server logic.phyloDB.git/docker/instance/db: files corresponding to the Neo4j database data.
The container files themselves are excluded from this repository through phyloDB.git/docker/.gitignore.
Again, you should be in directory phyloDB.git/docker.
./stop-docker.sh
A pristine Neo4j instance should be running for testing.
Unit tests (also in the CI workflow):
cd phylodb
./gradlew test --tests pt.ist.meic.phylodb.unit*
Performance tests:
cd phylodb
./gradlew test --tests pt.ist.meic.phylodb.performance.AlgorithmBenchmarks
./gradlew test --tests pt.ist.meic.phylodb.performance.OperationBenchmarks
Algorithm module tests (a Neo4j instance is not required):
cd algorithms
mvn test
An example is available at phyloDB.git/scripts/example.
The script example.sh illustrates the API use cases.
It is important that a Google account is registered as admin, and that you have a valid token.
Check the initialization step in the example.sh script, and/or the template for data initialization.
To run the example,
export TOKEN="YOUR_TOKEN"
export USER_ID="YOUR_USER_ID"
git clone https://github.com/phyloviz/phyloDB.git phyloDB.git
cd phyloDB.git/
sed 's|#spring.data.neo4j.uri=bolt://phylodb-neo4j:7687|spring.data.neo4j.uri=bolt://phylodb-neo4j:7687|g' -i phylodb/bin/main/application.properties
sed 's|spring.data.neo4j.uri=bolt://localhost:7687|#spring.data.neo4j.uri=bolt://localhost:7687|g' -i phylodb/bin/main/application.properties
./configure.sh
cd docker
./build-docker.sh
./start-docker.sh
cd ../scripts/example
echo y | ./example.sh
To generate YOUR_TOKEN, access https://developers.google.com/oauthplayground/, #, and click 'Authorize
APIs' with the custom scope https://www.googleapis.com/auth/userinfo.email.
Click in 'Exchange authorization code for tokens' and use the provided 'Access token'.
Use the account email associated with this token as YOUR_USER_ID.
The Wiki of this project documents several topics, namely archictecural views, deployment, authentication and API definition (usage).
- Bruno Lourenço (Developer)
- Miguel Coimbra (Developer, Maintainer)
- Cátia Vaz (Supervisor, Maintainer) (cvaz at cc.isel.ipl.pt)
- Alexandre Francisco (Supervisor Maintainer) (aplf at ist.utl.pt)