AbstractBackgroundAlzheimer’s Disease (AD) is a complex, heterogeneous, and multi‐factorial disease of predominantly the aging population. Throughout disease progression, different brain regions and cell types within each region show varying levels of resilience to ongoing disease pathology. A better understanding of the biological properties of different cell types and an integrated analysis approach that incorporates information about the interplay between cell type vulnerability, pathological deposits, and inflammatory signatures are required for the development of novel therapeutics that delay the onset or slow the progression of AD.MethodCerevance has developed a proprietary Nuclear Enriched Transcript Sort sequencing (NETS‐seq) method that provides deep molecular insight into the dynamics of the brains of early‐stage AD, late‐stage AD, and non‐neurodegenerative age‐matched control donors at the individual cell type level. Machine learning methodologies based on gene expression were employed to stratify donors across the disease continuum and to identify gene changes associated with disease progression.ResultNETS‐seq enabled us to profile more than a dozen distinct cortical cell types. These data show that many known AD risk genes, as well as many lesser described genes, change in expression in disease. We have also used pseudotime trajectory analysis to rank donors across disease progression from control to early‐ and late‐stage AD. Comparisons made across different cell types within the same brain region and the same cell types from different brain regions, have identified genes that may be contributing to AD progression at different stages of disease. These genes may contribute to ongoing neuronal vulnerability.ConclusionWe have produced highly reproducible molecular profiles from neuronal (and glial) cell types from the human brain and demonstrated that many genes are specifically expressed in cortical neuronal subtypes and change across the AD disease continuum. Identifying changes in the transcriptomic signatures of vulnerable neuronal populations at different stages of disease has enabled the identification and selection of novel targets for AD drug discovery.