Skip to content

For developers

Phenopacket Store repository presents collections of GA4GH phenopackets that can be used to test phenopacket-based software.

Each collection features phenopackets representing clinical information about individuals with Mendelian disease-associated variants in specific genes, or in some case, collections of phenopackets derived from published cohorts or studies.

pyphetools

Most of the phenopackets were created using the pyphetools library. Details about how the phenopackets were generated can be found in the notebooks folder. The folder is organized according to gene or collection. In some cases, the phenopackets were generated with other software, and then we present a notebook that summarizes the phenopackets.

Generate release archive

All phenopackets are stored in subfolders of the notebooks folder, which is organized according to cohort.

We use Phenopacket Store Toolkit (ppktstore) to package the phenopackets into a release ZIP file.

Run the following in order to create the release ZIP archive.

First, make sure to start in the top-level folder of the phenopacket store repository. Then, install phenopacket-store-toolkit in release profile into your active Python environment.

python3 -m pip install phenopacket-store-toolkit[release]

ppktstore comes with a command line interface (CLI) which includes the package command for generating the archive:

python3 -m ppktstore package --notebook-dir notebooks --release-tag 0.1.18 --output all_phenopackets

The command will collect the phenopacketes of the notebooks directory and pack the phenopackets into all_phenopackets.zip archive. We use 0.1.18 as the release tag.

Running the notebooks locally

As an alternative to using the phenopackets from one of the release archives, the phenopackets can also be generated locally by running Jupyter notebooks.

There are several ways of doing this and we prefer the following method. First, create a virtual environment (shown here as my_venv, but choose any name you like) and activate it. Then, install the following packages:

  • pyphetools required for generating phenopackets
  • ipykernel for using my_venv as a Jupyter kernel
# Create and activate the Python environment
ENV_NAME=my_venv
python3 -m venv ${ENV_NAME}
source ${ENV_NAME}/bin/activate

# Install both packages
python3 -m pip install pyphetools ipykernel

# Register the environment with Jupyter to use as a notebook kernel
python3 -m ipykernel install --user --name ${ENV_NAME} --display-name "Phenopacket Store Env"

After the installation, a new kernel called Phenopacket Store Env should be available in Jupyter. Use the kernel to run the notebooks at will Make sure to choose the kernel when running the notebooks in the notebooks folder.