This project is a web-based application for predicting drug-drug interactions (DDIs) using machine learning models. It integrates biomedical datasets (like TWOSIDES, DrugBank, and RDF2Vec embeddings) with pre-trained models (MLP, XGBoost) to identify potential adverse interactions between drugs.
- ✅ Built with Flask (Python)
- 🧬 Uses drug embeddings (
rdf2vec) and curated drug name mappings - 🤖 Pre-trained models:
MLP(.h5file)XGBoost(.pkland.json)
- 📊 Data: TWOSIDES, DrugBank, RxNorm mappings
- 🧪 Web interface for predicting and displaying interactions
git clone https://github.com/Logicrithm/DDI.git
cd DDI- Install Dependencies
Make sure you have Python 3.8+ installed. Then install required packages:
pip install -r requirements.txt- Run the Web App
cd app
python app.py