Alzheimer’s disease (AD) is a neurodegenerative disorder characterized by dysfunction and deterioration of neurons resulting in loss of memory and progressive cognitive decline. Current treatments are aimed only at symptom management. Three barriers to effective therapeutic development include: 1) a lack of definition of the heterogeneity of AD pathogenesis, 2) a lack of highly predictive biomarkers to facilitate early intervention, and 3) a need to identify pathways involved in cognitive decline and AD that can be targeted for therapeutic intervention. We are using induced pluripotent stem cell (iPSC) technology coupled to comprehensive studies of patient populations to interrogate the cellular and molecular mechanisms underlying AD in an effort to break down these barriers. We propose that there exist multiple forms of AD that have different underlying causes, and that multiple therapeutic interventions may be needed to address disparate etiologies. Through other funding sources, we have generated 50 iPSC lines from two cohorts, the Religious Order Study (ROS) and the Memory and Aging Project (MAP). Here, we propose to study these lines, where we focus upon three categories of subjects that lie on the extreme ends of the pathological spectrum: 1) no brain pathology, not cognitively impaired, 2) high pathology not cognitively impaired, and 3) high pathology, late onset Alzheimer’s disease. In addition, we have generated and/or collected iPSC lines from familial AD subjects, which will be analyzed in parallel. Using iPSC derived neurons, astrocytes, and microglia from 60 human subjects, we will measure AD relevant outcomes (Aβ, p-tau, cytokines/chemokines) in Aim 1 and acquire unbiased transcriptomic and proteomic data in Aim 2. These data will be integrated with clinical data, neuropathology data, and genetic data acquired from the same subjects from whom the cells were derived using multiple computational approaches. We hypothesize that: 1) Some pathological findings in the postmortem brain can be predicted by in vitro cellular assays on iPSC derived neurons and glia (Aim 1), and 2) Both neuropathology and cognitive decline in some human subjects can be predicted by transcriptomic and proteomic level network analyses of iPSC derived neurons and glia. Preliminary data from 12 lines supports the premise that iPSC-derived cells will capture certain cell and molecular signatures that define subgroups of aged adults. Data from all 50 ROS/MAP lines and EOAD models generated in aims 1 and 2 will provide a well-defined framework to address mechanistic questions regarding AD. In Aim 3 we will leverage the deeply characterized set of iPSC-derived cultures to address the hypothesis that dysregulated inositol 5-phosphatase activity of I...