Skip to content

About

OmicsGuard is a flexible, serverless-ready validator that supports any JSON schema validation, allowing for custom business logic and extensions beyond its built-in bioinformatics standards.

Topics

Resources

Stars

1 star

Watchers

0 watching

Forks

Repository files navigation

DOI

OmicsGuard

Python application PyPI version License: MIT Lines of Code

OmicsGuard is a lightweight, serverless-ready metadata validator designed for modern bioinformatics workflows. It ensures "Schema-on-Write" integrity for GA4GH standards (like Phenopackets) while allowing flexible organizational extensions and custom business logic.

Key Features

  • Serverless Ready: Designed to run in AWS Lambda, Google Cloud Functions, or lightweight containers.
  • GA4GH Standard Support: Built-in registry for official schemas (e.g., ga4gh-phenopacket-v2).
  • Plugin System: Inject custom business logic with simple Python functions.
  • Schema Extensions: Layer organization-specific fields (e.g., hospital_id) on top of standard schemas.
  • Human-Readable Reports: Generate HTML or Markdown reports for biologists and clinicians.
  • Smart Caching: Automatically caches remote schemas to minimize network latency.

Installation

You can install OmicsGuard from PyPI or directly from the source.

1. From PyPI (Recommended)

pip install omicsguard

2. From Source

Useful for development or using the latest unreleased features.

git clone https://github.com/your-org/omicsguard.git
cd omicsguard
pip install .

Usage

1. Basic Validation

Validate a JSON or YAML file against the default bundled standard (Phenopackets v2).

omicsguard validate --data patient_data.json

Output:

true

2. Pulling Schemas

Download and cache official schemas from the registry to avoid repeated network calls.

omicsguard pull --standard ga4gh-phenopacket-v2

3. Custom Business Rules (Plugins)

Sometimes, structural validation isn't enough. Use Python plugins to enforce complex rules (e.g., "ID must start with PROBAND-").

Create a plugin (my_plugin.py):

def validate(data):
    errors = []
    if not data.get("id", "").startswith("PROBAND-"):
        errors.append("ID must start with 'PROBAND-'")
    return errors

Run with plugin:

omicsguard validate --data patient.json --plugin my_plugin.py

4. Schema Extensions

Add custom fields to a standard schema without forking it.

Extension Schema (hospital_fields.json):

{
  "properties": {
    "hospital_id": { "type": "string" }
  },
  "required": ["hospital_id"]
}

Merge and Validate:

omicsguard validate --data patient.json --extend hospital_fields.json

5. Generating Reports

Generate detailed reports for non-technical stakeholders.

# HTML Report
omicsguard validate --data patient.json --output report.html

# Markdown Report (great for CI/CD comments)
omicsguard validate --data patient.json --output report.md

Configuration

OmicsGuard is 12-factor app compliant and can be configured via Environment Variables:

  • OMICSGUARD_SCHEMA_CACHE: Directory to store cached schemas (Default: ~/.omicsguard/schemas).

Development

  1. Clone the repo

    git clone https://github.com/your-org/omicsguard.git
    cd omicsguard
  2. Install dependencies

    pip install -e .
  3. Run Tests

    python -m unittest discover tests

License

MIT License. See LICENSE for details.

About

OmicsGuard is a flexible, serverless-ready validator that supports any JSON schema validation, allowing for custom business logic and extensions beyond its built-in bioinformatics standards.

Topics

Resources

Stars

1 star

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages