AMPSphereBackend

This repository contains source code for the backend of AMPSphere website

Deploy instructions

Prepare necessary data

Copy the latest AMPSphere data (should be generated based on Celio's script) to data/original_data. The folder will be look like.

data/original_data
├── AMPSphere_generation_v.2022-03
│   ├── analysis
│   ├── data
│   ├── main.py
│   ├── README.md
│   ├── seqlogo
│   └── utils
├── metadata_analysis
│   ├── data
│   ├── main.py
│   ├── outputs
│   ├── README.md
│   └── utils
└── zenodo_repo
    ├── AMPSphere_v.2022-03.faa.gz
    ├── AMPSphere_v.2022-03.fna.xz
    ├── AMPSphere_v.2022-03.general_geneinfo.tsv.gz
    ├── AMPSphere_v.2022-03.quality_assessment.tsv.gz
    ├── README.md
    └── SPHERE_v.2022-03.levels_assessment.tsv.gz

Prepare the environment

conda install -c default -c conda-forge -c bioconda \
    fastapi gunicorn sqlalchemy pandas pytest pydantic biopython requests
    mmseqs2 hmmer uvicorn

# For testing
conda install -c default -c conda-forge httpx

Generate the backend data and test on dev server

Important: Before executing the data generation step, make sure you have the right VERSION_CODE argument in the script (edit it to modify).

./generate_data.sh  
python -m pytest tests

This will cost hours to half a day depends on the available computing resources.

Transfer all the data to the production server

tar zcvf data.tgz --exclude "data/original_data" data
scp data.tgz ampsphere-api.big-data-biology.org:/AMPSphere/website/AMPSphereBackend/

Uncompress the data and do unittest before production

tar zxvf data.tgz
python -m pytest tests/

If everything is Okay, please run the backend in production mode.

Go to its API page and test it manually if necessary: https://ampsphere-api.big-data-biology.org/

Contact

• Luis Pedro Coelho