Pharma-AI - Biomedical Pharmacognosy Agent

An AI-powered biomedical agent specialized in pharmacognosy.

---

What is Pharma-AI?

Pharma-AI is an intelligent biomedical assistant that helps researchers, pharmacists, and healthcare professionals explore medicinal plants, their active compounds, and potential drug interactions. It combines retrieval-augmented generation (RAG) with a specialized LLM to provide evidence-based answers grounded in scientific literature.

Core use case: type the name of a medicinal plant and get active compounds, mechanisms of action, scientific references, and drug interaction warnings.

---

---

Features

• AI Agent (RAG + Llama 3) for biomedical Q&A
• Drug Interaction Checker (plant-drug and plant-plant)
• Scientific grounding with PubChem and PubMed/NCBI data
• Streamlit web interface for practical exploration

---

Project Structure

Pharma-AI/
|-- pubchem_data.py        # Fetches active compounds from PubChem API
|-- pubmed_data.py         # Fetches scientific articles from PubMed/NCBI
|-- rag_knowledge.py       # Builds the FAISS vector knowledge base
|-- interactions_db.py     # Database of known plant-drug interactions
|-- interactions.py        # Interaction detection engine
|-- agent.py               # RAG agent via Groq API
|-- app.py                 # Streamlit web interface
|-- .env                   # API keys (never commit)
|-- .gitignore
`-- requirements.txt

---

Quick Start

1. Clone the repository

git clone https://github.com/your-username/Pharma-AI.git
cd Pharma-AI

2. Create virtual environment

python -m venv .venv

# Windows
.venv\Scripts\activate

# Mac / Linux
source .venv/bin/activate

3. Install dependencies

pip install -r requirements.txt

4. Configure environment variables

Create a .env file at the project root:

GROQ_API_KEY=your_groq_api_key_here
ADMIN_EMAIL=your_admin_email
ADMIN_PASSWORD=your_admin_password
SESSION_SECRET=a_long_random_secret

Get a free Groq API key at console.groq.com.

5. Build the knowledge base

python pubchem_data.py
python pubmed_data.py
python rag_knowledge.py

6. Launch the app

streamlit run app.py

Open http://localhost:8501 in your browser.

---

Requirements

• langchain
• langchain-community
• langchain-groq
• langchain-huggingface
• langchain-text-splitters
• faiss-cpu
• sentence-transformers
• streamlit
• requests
• pandas
• biopython
• python-dotenv
• groq

---

Data Sources

• PubChem: molecular compounds and properties
• PubMed / NCBI: scientific abstracts and articles
• Groq / Llama 3.1: LLM inference

---

Security

• Never commit your .env file.
• If a key is exposed, revoke it immediately.
• Use Streamlit Secrets for production deployments.

---

License

MIT License.