A multimodal brain cell atlas and community resource of Alzheimer's disease and comorbid dementias.

Single cell and spatial genomics methods that can be used to study the human brain are rapidly increasing our understanding of the fine cellular and molecular organization of the neurotypical brain and are being catalyzed by the NIH BRAIN Initiative. The first phase of our NIA U19 Center called the Seattle Alzheimer’s Disease Brain Cell Atlas (SEA-AD) was designed to adapt these technologies to study postmortem tissues across the spectrum of AD pathology, establish the use of these data to identify vulnerable and affected cell types and dysregulated transcriptional and epigenetic patterning as a function of increasing disease severity and across affected brain regions, and create an openly accessible data resource. SEA-AD has successfully established this paradigm, producing the largest, deepest coverage multimodal datasets in the field to date, and demonstrated its effectiveness at modeling disease trajectories and identifying the earliest affected cell types and molecular patterns. This approach is now dramatically empowered by new advances from the next phase of the BRAIN Initiative Cell Atlas Network (BICAN), which is developing new technologies and creating foundational brain-wide human brain cell atlases that can be used as normative references in the study of AD. The current proposal (SEA-AD2.0) aims to leverage and scale this paradigm to create an atlas of AD and comorbid dementias (AD/ADRD) and a community hub to make these data and technologies broadly available to the community. This atlas will describe cellular, spatial, transcriptomic, and epigenomic changes that characterize the progression of AD across the brain and over the course of disease. A major focus is to map changes over the brain in a large highly heterogeneous cohort that captures both AD and AD/ADRD polypathologies, genetic backgrounds, topological and cognitive subtypes, and aged donors resistant to pathology or having AD pathology but cognitively resilient. To assemble the necessary high quality donor cohorts, BICAN advances in brain preparation for tissue quality will be implemented across a network of AD tissue resources with different clinical cohort demographics, thus allowing mapping into fMRI-based templates to increase anatomical precision. The core SEA-AD strategy of quantitative neuropathology, single nucleus Multiome (combined RNA-seq and ATAC-seq), and spatial transcriptomics will be used to analyze tissues across functionally important brain regions across the spectrum of AD and AD/ADRD pathology. Furthermore, the project will establish an Integrated Analysis Center to foster joint analysis across single cell transcriptomics efforts across the field, by using mapping against the SEA-AD cell classification as way to harmonize and jointly analyze data from multiple major AD efforts. The resulting data will be used to model disease pseudo-progression, identify vulnerable and affected cell types, identify differences among AD subtypes, and identify hallmarks of disease common across this wide range of disease heterogeneity or specific to certain subsets of the cohort. A key philosophy for SEA-AD2.0 is open science, with a major emphasis on pre-publication data release and expansion of data resource capabilities, particularly through tight linkage with the BICAN Brain Knowledge Platform, a resource for spatial and annotated cellular references, as well as key AD resources and data visualization tools for useful community access.