Integrating AI-enhanced kinase enrichment analysis (KEA) with geometric deep learning and federated learning for precision drug repurposing
Abstract
Artificial intelligence (AI) is reshaping drug repurposing by integrating systems biology with molecular design. Here, we present a unified framework combining AI-enhanced Kinase Enrichment Analysis (KEA), geometric deep learning, and federated learning to enable scalable and privacy-preserving therapeutic discovery. KEA prioritizes disease-relevant kinases from multi-omics data, while geometric deep learning captures structure-activity relationships (SARs) at atomic resolution. Federated learning facilitates secure, multi-institutional model training across heterogeneous datasets. This integrative pipeline enhances identification of repurposable kinase inhibitors and supports emerging modalities, such as proteolysis-targeting chimeras (PROTACs). A case study in Alzheimer's disease (AD) highlights improved target prioritization and predictive performance. By bridging kinase signaling networks with AI-driven modeling, this framework provides a robust strategy for accelerating precision drug discovery and repurposing.
Read the paper
Related studies
- Active Learning during Federated Learning for Object Detection
- Fairness-aware federated learning
- Decentralized Federated Learning
- Building a Personally Identifiable Information Recognizer in a Privacy Preserved Manner Using Automated Annotation and Federated Learning
- Preservation of Patient Level Privacy: Federated Classification and Calibration Models
- Efficient explainable federated learning
Go deeper in Evidence Hub
Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.