https://github.com/deepbiolab/rag-for-proteins

A demonstration of Retrieval-Augmented Generation (RAG) applied to protein analysis
https://github.com/deepbiolab/rag-for-proteins

llm local-rag proteins rag

Last synced: 3 months ago
JSON representation

A demonstration of Retrieval-Augmented Generation (RAG) applied to protein analysis

• Host: GitHub
• URL: https://github.com/deepbiolab/rag-for-proteins
• Owner: deepbiolab
• License: mit
• Created: 2025-04-06T12:39:50.000Z (over 1 year ago)
• Default Branch: main
• Last Pushed: 2025-04-07T15:59:03.000Z (about 1 year ago)
• Last Synced: 2025-06-04T08:10:49.152Z (about 1 year ago)
• Topics: llm, local-rag, proteins, rag
• Language: Python
• Homepage:
• Size: 347 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# RAG for Proteins

![](./assets/rag.png)

A demonstration of Retrieval-Augmented Generation (RAG) applied to protein analysis, currently focusing on antibody data from SAbDab (Structural Antibody Database).

## 🌟 Features

Current:

- RAG-powered chat interface for antibody structure analysis

- Integration with SAbDab for antibody structural data

- Local LLM support via Ollama (currently using Qwen 7B)

- Persistent vector storage for efficient retrieval

- Clean, modular architecture with separate components for:

  - Data loading and preprocessing

  - Vector storage management

  - LLM interface

  - RAG pipeline coordination

  - Streamlit UI

## 🚀 Quick Start

### Prerequisites

- Python 3.10+

- Ollama installed and running

- Required Python packages:

```bash

pip install -r requirements.txt