Metadata-Version: 2.4
Name: bsbt
Version: 0.1.1
Summary: BSBT: Binding Sequence Benchmarking Tool - A Python package for benchmarking protein language models and predicting antibody-antigen binding.
Author-email: Ayush Laware <ayushlaware@gmail.com>
License: MIT
Project-URL: Homepage, https://github.com/ayushlaware/BSBT
Requires-Python: >=3.8
Description-Content-Type: text/markdown
Requires-Dist: torch
Requires-Dist: transformers
Requires-Dist: scikit-learn
Requires-Dist: numpy
Requires-Dist: pandas

# 🧬 BSBT – Antibody to Antigen Prediction

BSBT (Biological Sequence Binding Tool) is a Python package designed to **predict antigen sequences given antibody sequences**.  
The package is structured to allow easy experimentation and integration of **protein language models (PLMs)** or **large language models (LLMs)** in the future.

At the moment, BSBT provides a **mock implementation** using hardcoded antibody–antigen sequence mappings. This enables developers and researchers to test the pipeline structure before connecting it to real machine learning models.

---

## ✨ Features

- 🔹 Simple **antibody → antigen prediction** interface (`findAntigen`)
- 🔹 Current backend: **mock hardcoded mappings**
- 🔹 Future backend: **LLM or fine-tuned PLM integration**
- 🔹 Modular design (easy to extend, swap components)
- 🔹 Unit tests included (`pytest`-ready)

---

## 📦 Installation

You can install directly from PyPI (after publishing):

```bash
pip install bsbt
