Getting started

The gocam package provides Pydantic models and a command-line interface for working with Gene Ontology Causal Activity Models (GO-CAMs). See the GO-CAM overview for background on the model and use the GO-CAM Browser to explore published models.

Installation

Install the package from PyPI:

pip install gocam

This installs both the Python package and the gocam command.

Fetch a published model from the command line

The fetch command retrieves a published model and translates it into the package's GO-CAM data model. YAML is the default output format:

gocam fetch 5b91dbd100002057 > model.yaml

Request JSON with --format:

gocam fetch --format json 5b91dbd100002057 > model.json

Run gocam fetch --help for all fetch options.

Fetch a published model from Python

MinervaWrapper fetches the published Minerva representation and translates it into a validated Model:

from gocam.translation import MinervaWrapper

model = MinervaWrapper().fetch_model("5b91dbd100002057")
print(model.id)
print(model.title)

Load a local model

Loading a local file validates it against the Pydantic model included in the installed gocam version.

JSON

from pathlib import Path

from gocam.datamodel import Model

model = Model.model_validate_json(Path("model.json").read_text())
print(model.id)

YAML

from pathlib import Path

import yaml

from gocam.datamodel import Model

with Path("model.yaml").open() as stream:
    model = Model.model_validate(yaml.safe_load(stream))

print(model.id)

Work with a model

The validated object exposes model fields as Python attributes:

print(model.title)
print(model.taxon)
print(len(model.activities or []))

Convert it back to JSON-compatible Python values with Pydantic:

model_data = model.model_dump(mode="json", exclude_none=True)

See the schema reference for all model classes and fields. For CX2, gene-to-gene, and NetworkX workflows, continue to Advanced translation.