ABSTRACT Large multi-center GWA studies have found associations between over 20 genomic loci and late-onset Alzheimer’s disease (LOAD). However, the precise target genes, the causal genetic variants and their molecular mechanisms of action through which they exert their pathogenic effects remain largely unknown. Our long-term goal is to elucidate causal genetic factors and their functional effects that contribute to the risk of developing LOAD. Our central hypothesis is that changes in expression levels of critical disease genes is an important molecular mechanism underlying LOAD etiology and that causal variants modulate expression of these disease genes, and by that contribute to LOAD risk. Changes in gene expression in LOAD vs. healthy controls were described in brain tissues by our team and others and previous studies reported the cis- associations of tagging SNPs with expression of nearby LOAD-risk genes, providing a strong scientific premise for the proposed study. In this study, we will employ a multifaceted approach that combines in silico, in vitro and in vivo methods to investigate regions in the genome that were significantly associated with LOAD-risk in GWA studies. In Aim 1 we will identify target genes within LOAD-associated regions that show differential expression along the neuropathological progression of LOAD. We will determine the expression profile of genes within these regions in neurons, astrocytes and microglia isolated from affected and unaffected rapidly autopsied human brain tissues using laser capture microdissection (LCM) coupled with nCounter single cell gene expression technology (NanoString). Aim 2 will discover regulatory noncoding sequences within LOAD- associated regions. First, we will prioritize candidate regulatory elements using bioinformatics tools and human genome databases, as well as ATAC-sequencing experiments using NeuN+/- nuclei from affected and unaffected human brain tissues to determine chromatin accessibility profiles in cell type- and pathological stage- specific manners. The functionality of the candidate noncoding sequences will be then characterized using iPSC-derived model systems that will be genome edited to carry deletions of the predicted regulatory sequences. Aim 3 will focus on Short Structural Variants (SSVs) and will investigate the functional effects and causality of SSVs in the candidate regulatory sequences. We will use SMRT sequencing combined with Cas9 system (PacBio) to accurately determine the SSVs genotype and haplotypes in LOAD compared to control subjects, and will examine their regulatory effects using genome edited isogenic iPSC-derived neurons and/or astrocytes models that carry different alleles/haplotypes at the SSV site. Our study will advance the identification of causal genetic factors and the understanding of their molecular effects that contribute to the risk of developing LOAD. This knowledge will provide insight regarding ...