PROJECT SUMMARY Although researchers have conducted more than 400 human trials for potential treatments of Alzheimer’s disease (AD) in the last two decades, the attrition rate is estimated at over 99%. Furthermore, the “one gene, one drug, one disease” reductionism-informed paradigm overlooks the inherent complexity of the disease and continues to challenge drug discovery for AD. The predisposition to AD involves a complex, polygenic, and pleiotropic genetic architecture. Recent studies have suggested that AD often has common underlying mechanisms and pathobiology, sharing intermediate endophenotypes with many other complex diseases. These endophenotypes, such as amyloidosis, tauopathy and neuroinflammation, have essential roles in many neurodegenerative diseases. Systematic identification and characterization of novel underlying pathogenesis and endophenotype networks, more so than mutated genes, will serve as a foundation for generating actionable targets as input for drug repurposing and rational design of combination therapy in AD. Integration of the genome, transcriptome, proteome, and the human interactome using artificial intelligence (AI) and machine learning (ML) are essential for such identification. Given our preliminary results, we posit that AI/ML-based identification of likely risk genes and endophenotype network modules offer unexpected opportunities for drug repurposing and combination therapy design in AD compared to traditional single-target approaches. To address the underlying hypothesis, we propose to establish an AI/ML-based, multimodal analytic framework to repurpose existing genetics, genomics and transcriptomics data generated from NIA-funded AD genome sequencing projects for druggable target identification with two specific aims under the scope of the parent R01 (#R01AG066707). The central unifying hypothesis of this Supplement project is that a genome-wide, AI/ML infrastructure that enables users searching, sharing, visualizing, querying, and analyzing multi-omics (including genetics and genomics) findings can enable emerging development of molecularly targeted treatments for AD. Aim 1 will test common variant-based risk gene and endophenotype network hypothesis in AD using multi-omics evidence aggregation under a multiple kernel learning framework and the FAIR (Findable, Accessible, Interoperable, and Reusable digital objects) principles. We will develop and apply AI/ML approach to identify likely risk genes and endophenotype networks though leveraging genetic, genomic, transcriptomic, and clinical data from AD Sequencing Project (ADSP), the AD Neuroimaging Initiative (ADNI), NIAGADS, and the AD knowledge portal. Aim 2 will test cell type-specific risk genes and anti-inflammatory endophenotype network hypothesis in AD using a network-based deep learning framework. Following FAIR principles, we will implement command-line and web portal to disseminate all AI/ML toolboxes and AI/ML-ready gene/network data from Aims 1 and...