🧬 OncoTrace.ai

An AI-powered web application designed to identify potential cancer-associated biomarkers. This project leverages a NextJS frontend, a Python (Flask + Streamlit) backend, and a trained machine learning model to analyze input gene data and return relevant biomarker predictions.

---

📚 Table of Contents

• 🧾 Project Description
• ⚙️ Tech Stack
• 🚀 Getting Started
  • 1️⃣ Clone the Repository
  • 2️⃣ Set Up the Backend
  • 3️⃣ Run the Frontend
  • 4️⃣ Start All Services
• 📦 Prerequisites

---

🧾 Project Description

This application offers an intuitive interface for identifying cancer-associated biomarkers. It combines biomedical NLP techniques with deep learning models to analyze gene data and surface insights using a web interface.

---

⚙️ Tech Stack

• Frontend: React, Tailwind CSS
• Backend: Flask, Streamlit, Python
• ML Models: BioBERT, OncoKB integration
• Other Tools: pandas, scikit-learn, torch

---

🚀 Getting Started

1️⃣ Clone the Repository

git clone https://github.com/AnishMane/Cancer-Associated-Biomarker-Identification.git
cd Cancer-Associated-Biomarker-Identification

---

2️⃣ Set Up the Backend

cd backend
python -m venv venv         # Use `python3` if needed
venv/Scripts/activate       # or `source venv/bin/activate` on macOS/Linux
pip install -r requirements.txt

---

3️⃣ Run the Frontend

cd frontend
npm install
npm run dev

---

4️⃣ Start All Services

Ensure you are in the backend folder and your virtual environment is activated:

# In one terminal (inside backend/)
python streamlit_app.py

# In a second terminal (inside backend/)
python inference.py

# In a third terminal (inside backend/)
python app.py

Your frontend should now be running on http://localhost:5173 and backend services (APIs and Streamlit dashboard) should be active.

---

📦 Prerequisites

• Node.js & npm – Download here
• Python 3.x & pip
• virtualenv (optional but recommended)