- This topic has 3 replies, 2 voices, and was last updated 2 months, 2 weeks ago by
.
-
Posts
-
KEDD (Knowledge-Enhanced Drug Discovery) is an AI-powered framework that integrates multiple knowledge modalities for AI-driven drug discovery. Unlike conventional models that rely solely on chemical structures, KEDD incorporates biomedical text, molecular graphs, protein sequences, and pharmacological data, creating a comprehensive AI system for identifying and optimizing drug candidates. It enables researchers to combine heterogeneous data sources to improve drug-target interaction prediction, molecular generation, and lead optimization.
Key Features:
Multi-Modal Knowledge Integration: Merges biomedical literature, molecular graphs, and protein data for drug discovery.
Enhanced AI-Driven Drug Design: Optimized for lead identification, toxicity prediction, and pharmacokinetics modeling.
Cross-Modal Learning: Supports drug discovery pipelines that integrate textual, chemical, and structural data.
Applications: AI-assisted drug repurposing, precision medicine, computational chemistry, and biomolecular interaction modeling.Publication:
📄 KEDD: Towards Unified AI Drug Discovery with Multiple Knowledge Modalities
🔗 Read the PaperThis forum topic is for computational chemists, bioinformaticians, and AI researchers working with KEDD. Join this discussion to:
Explore how KEDD integrates multiple data sources for AI-driven drug discovery.
Discuss the advantages of combining biomedical literature, molecular graphs, and protein data.
Share real-world applications in AI-assisted drug repurposing and precision medicine.
Collaborate on optimizing KEDD for molecular interaction and pharmacological modeling.
🚀 Join us in advancing AI-powered drug discovery with KEDD!
- You must be logged in to reply to this topic.