Galaxy Tools Fetcher using BioBlend

This Python script connects to a Galaxy instance using the BioBlend library. It retrieves available tools and outputs their information in JSON format. The connection credentials (URL and API key) are securely managed using a .env file.

---

Features

✅ Connect to a Galaxy instance
✅ Retrieve and list available tools
✅ Securely manage API credentials via environment variables
✅ Extendable to use SBERT for further processing

---

Prerequisites

• Python 3.7+
• Access to a Galaxy instance with an API key

---

Installation

1. Clone this repository

   git clone https://github.com/johnnas12/galaxy_tool_suggestion.git
   cd galaxy_tool_suggestion

2. Create and activate a virtual environment (optional but recommended)

   python -m venv venv
   source venv/bin/activate    # On Windows, use venv\Scripts\activate
   

3. Install the required packages

   pip install -r requirements.txt
   

4. Create a .env file in the project root directory.

   GALAXY_URL=https://usegalaxy.org
   GALAXY_API_KEY=your_api_key_here
   
   

5. Usage

   python fetch_tools.py
   

Galaxy Tools Suggestion Using intfloat/e5-large

A FastAPI-powered AI agent for suggesting relevant Galaxy tools based on natural language queries using sentence embeddings and cosine similarity.

Usage of the agent

After cloning this repository go to agents

cd agents

📌 1️⃣ Configure Paths

Open configs/config.py and set the following paths:

from pydantic_settings import BaseSettings

class Settings(BaseSettings):
    data_to_encode_path: str = "data/tools_metadata.json"  # raw data (JSON format)
    embeddings_path: str = "embeddings/galaxy_embeddings.npy"
    metadata_path: str = "embeddings/galaxy_metadata.json"

📌 2️⃣ Encode Your Data

After setting your config, run the embedding script to encode your tools' metadata JSON into embeddings (.npy) and save metadata as a JSON file for lookup.

python agents/text_embedding.py

This will:

• Load your raw JSON metadata
• Generate sentence embeddings
• Save embeddings to embeddings/galaxy_embeddings.npy
• Save clean metadata to embeddings/galaxy_metadata.json

📌 3️⃣ Run the FastAPI Backend

Start your API server with:

uvicorn app:app --reload

📌 4️⃣ Test the API

🔍 Health Check

GET http://127.0.0.1:8000/health

Response

{"status": "ok"}

🎯 Tool Suggestion

POST http://127.0.0.1:8000/suggest

Request Body

{
  "query": "align sequences to a reference genome",
  "top_k": 5
}

Response

{
  "results": [
    {
      "name": "Tool A",
      "description": "Performs sequence alignment",
      "help": "long help section",
      "category": "Alignment",
      "score": 0.92
    }
  ]
}