Project Abstract Alzheimer’s disease (AD) affects about 10% of the US population of age 65 and up, and roughly 40 million people worldwide. Despite decades of studies, the disease etiology of AD still remains unclear, and till today, there is no curative treatment for AD. Typically, AD patient brains show characteristic neurofibrillary tangles (NFTs), which are composed of aggregated bundles of microtubule-associated protein tau in its truncated or hyperphosphorylated state, and the degree of NFTs in the brain usually positively correlates with the disease progression and cognitive decline. However, there also exists subpopulations of patients whose disease trajectories do not follow the typical NFT accumulation way. A subgroup of patients’ brain samples showed AD-like high NFTs but with no or low cognitive deterioration, while another subgroup of patients presents severe cognitive impairment but low NFT pathology. Studying these atypical AD subtypes and identifying key factors in such uncoupling between NFT pathology and cognitive impairment will not only improve precision medicine in AD, but also provide valuable information on how to prevent and slow down cognitive deterioration during the long disease progression. In this project, our overarching hypothesis is that the transcriptomic, proteomic, and network-level differences identified for the atypical NFTs-dementia uncoupled human brains can contribute to the mechanisms of rapid cognitive deterioration in low-NFT AD groups and the dementia resilience in Asymptomatic AD groups. We plan to apply data-driven approaches with integrative genomic analyses on multiple AD proteomic and transcriptomic datasets to identify candidate genes, proteins, and coexpression networks that play important roles in the NFTs-dementia uncoupling in the atypical AD patients. The identified gene candidates will be further screened in the Drosophila strains expressing human pathological tau for causal genes and proteins that are key factors for either NTF-dementia uncoupling or rendering neuron protection/susceptibility to AD. Promising candidates obtained from Drosophila experiments will be further tested in the mouse strains which can express human tau as early as four months old to understand their roles in terms of neuron-protection or AD-susceptibility. Our long- term goal is to generate experimental evidence for further grant applications to elucidate the NFT-AD uncoupling mechanisms behind the two subgroups, which can open new research directions for AD prevention as well as AD drug development.