Abstract Transcriptomic studies in clinical and biomedical research have mainly focused on changes in linear transcripts to provide knowledge of the genes and co-expression networks implicated in the disease. However, very little is known about the role of circular RNAs (circRNAs) in Alzheimer's disease (AD). CircRNAs are a novel category of non-coding RNAs derived from the back-splicing and covalent joining of pre-mRNA exons and introns. We recently performed a transcriptome-wide analysis of circRNA differential expression in the brain cortex from more than 621 brain samples from two independent and large cohorts of late-onset sporadic AD cases and neuropathology-free individuals. We identified specific circRNAs, including circHOMER1, associated with AD risk and neuropathological traits. This project will use brain tissue to identify additional brain circRNAs implicated in AD from a cohort that is four-fold larger than the cohort in our previous study. We also plan to investigate differentially expressed blood circRNAs in AD cases compared with controls to determine their biomarker utility for creating new prediction models. We will establish a framework for in vitro and in vivo functional characterization of the role of circRNAs in AD. As proof of principle, we will start by defining the role of circHOMER1 in AD-related gene expression and related cellular phenotypes. We have found that circHOMER1 is highly expressed in induced pluripotent stem cell (iPSC)-derived neurons. We will use CRISPR/Cas9 to knock down circHOMER1 as well as use circHOMER1 overexpression (OE) in iPSC-derived neurons from isogenic controls and AD patient-derived neuronal cultures from pathogenic mutation carriers of APP, PSEN1, PSEN2, and MAPT genes. To determine whether circHomer1 accelerate/increase AD-related pathology in vivo, we will generate circHomer1-knockout transgenic mice and cross them with 5XFAD and MAPT-P301S mice. We will also use AAV2/9-mediated circHomer1 OE in the cortex and hippocampus of 5XFAD and MAPT-P301S mice. We recently found that seven-month-old 5XFAD mice display significant reductions of circHomer1 expression, similar to the postmortem brains of AD patients. We will restore circHomer1 levels in the hippocampus and cortex of 5XFAD and P301S mice using AAV2/9 vectors. This proposal will be the first to systematically analyze the role of brain and blood circRNAs in AD and to perform in vitro and in vivo functional studies to characterize the role of circRNAs in AD.