Deployment

This section covers how to deploy Phentrieve in various environments, from local development to production systems.

Deployment Options

Phentrieve can be deployed in several ways, depending on your needs:

1. Local Installation: Install the Python package locally for development or personal use
2. Docker Containers: Deploy as containerized services for easier management and isolation
3. Production Deployment: Set up a full production environment with proper networking and security

Components

A complete Phentrieve deployment consists of:

• Core Python Package: The foundation that provides all functionality
• API Server: A FastAPI-based service that exposes endpoints for querying and processing
• Frontend: A Vue.js-based web interface for user-friendly interaction
• Vector Database: ChromaDB for storing and retrieving vector embeddings
• Data Storage: Directories for HPO data, indexes, and results

Section Contents

• Docker & NPM: Setting up Phentrieve using Docker and NPM for frontend development
• Local Docker: Running Phentrieve locally with Docker Compose
• Data Management: Managing Phentrieve data in different deployment scenarios