Integration of artificial intelligence and drug discovery to treat Alzheimer’s disease

About This Grant

Project Summary/Abstract Alzheimer’s disease (AD) is a devastating neurodegenerative disorder affecting approximately 44 million people worldwide. Due to the lack of a clear understanding of its underlying mechanisms, hypothesis-driven AD drug discovery remains a high-risk challenge. The NIA/NIH-supported Accelerating Medicines Partnership in Alzheimer’s Disease (AMP-AD) initiative has made significant progress in identifying and prioritizing a comprehensive list of novel AD-associated targets. However, the discovery of viable chemical probes for these targets remains difficult due to the inherent complexity of traditional screening methods and the challenging nature of these targets. Recent advancements in AI-driven computational methodologies have the potential to transform drug discovery by enabling more efficient hit identification and optimization. However, even the most sophisticated computational approaches are ineffective without seamless integration into a well-defined experimental pipeline. Without rigorous experimental validation and iterative refinement, AI-driven methods risk failing to translate their predictions into meaningful therapeutic outcomes. To address this limitation, we will develop and apply a tightly integrated AI-driven computational and experimental workflow to accelerate the discovery of small-molecule modulators of proteins well known to influence Alzheimer’s disease through the regulation of phagocytosis by microglia. Specifically, we will screen seven sites across four proteins: PLC2, moesin, GEF-H1, and SHIP1. Our approach will involve: 1) developing and apply AI-driven methods to design DNA-encoded libraries that target these sites; 2) implementing an integrative AI-driven computational and experimental pipeline for hit discovery and validation; and 3) performing AI-assisted hit optimization to enhance the efficacy of promising leads in cellular and animal models. Through this approach, we anticipate delivering potent small-molecule leads for the proposed targets while also developing and openly sharing highly innovative, rigorously validated AI-driven methodologies to advance drug discovery to treat Alzheimer’s disease.