PROJECT SUMMARY Predisposition to AD involves a complex, polygenic, and pleiotropic genetic architecture; furthermore, there are no disease modifying treatments that slow the neurodegenerative process for AD. Traditional reductionist paradigms overlook the inherent complexity of AD and have often led to treatments that are lack of clinical benefits or fraught with adverse effects. Existing multi-omics data resources, including genetics, genomics, transcriptomics, interactomics (protein-protein interactions and chromatin interactions), have not yet been fully utilized and integrated to explore the pathobiology and drug discovery for AD. Understanding AD genetics and genomics from the point-of-view of how cellular systems and molecular interactome perturbations underlie the disease (termed disease module) is the essence of network medicine. Systematic identification and characterization of novel underlying pathogenesis and disease module, will serve as a foundation for identifying and validating novel risk genes and drug targets in AD. Given our preliminary results, we posit that a genome- wide, multimodal artificial intelligence (AI) framework to identify new risk genes and networks from human genome/exome sequencing and multi-omics findings enable a more complete mechanistic understanding of AD pathogenesis and the rapid development of targeted therapeutic intervention for AD with great success. Aim 1 will determine whether rare coding and non-coding variants by whole-genome/exome sequencing (WGS/WES) are enriched in protein-functional and gene-regulatory regions using sequence and structure-based deep learning models. These analyses will assemble WGS/WES and clinical data from Alzheimer's Disease Sequencing Project (ADSP), publicly available protein structure (i.e., protein-protein interfaces, protein-ligand binding sites, post-translational modifications) and sequence (expression quantitative trait locus [eQTLs], histone-QTLs, and transcription factor binding-QTLs) information from the PDB database, GTEx, NIH RoadMap, FANTOM5, PsychENCODE, and NIH 4D Nucleome. Aim 2 will determine whether GWAS common variants linked to AD pathobiology and endophenotypes are enriched in gene regulatory networks in a cell-type specific manner using a Bayesian framework. We will validate risk gene and network findings using WGS/WES and protein panel expression data from our existing cohorts: The Cleveland Clinic Lou Ruvo Center for Brain Health Aging and Neurodegenerative Disease Biobank (CBH-Biobank) and the Cleveland Alzheimer's Disease Research Center (CADRC). Aim 3 will test the hypothesis that risk genes and networks can be modulated via in silico drug repurposing, population-based validation, and functional test, to identify candidate agents and drug combinations that will modify AD. The successful completion of this project will offer capable and intelligent computer-based toolboxes that enable searching, sharing, visualizing, querying, and analyzing genetics, genomics, and multi-omics profiling data for genome-informed therapeutic discoveries for AD and other neurodegenerative disease if broadly applied.

Project Narrative It is estimated that more than 16 million people with AD live in the United States by 2050 and the predisposition to AD involves a complex, polygenic, and pleiotropic genetic architecture. This proposal will develop intelligent computer-based network medicine and systems biology tools, capable of identifying and validating human genome sequencing findings for novel risk gene discoveries and targeted therapeutic development in AD. The innovative network-based, artificial intelligence toolboxes and novel risk genes and biologically relevant targeted therapeutic approaches developed in this proposal will prove to be novel and effective ways to improve outcomes in long-term brain care for the rapidly growing AD population, an essential goal of AD precision medicine.