The lab develops interdisciplinary methods for AI-driven drug design. We focus on key challenges in early-stage drug discovery, including molecular activity, drug-likeness, safety, and developability. By integrating medicinal chemistry knowledge, machine learning models, and multimodal biomedical data, the lab aims to build reliable, efficient, and interpretable AI models that support molecular property prediction, molecular optimization, and data-driven design decisions.
Current research directions include:
- AI-based molecular property prediction
- Multi-property molecular optimization
- Explainable AI for drug design