ABSTRACT With an aging population, the impact of Alzheimer's disease (AD) on public health continues to explode. Altered daily rhythms in physiology and behavior are prominent features of AD. These altered activity rhythms are difficult to treat, disrupting the lives of both patients and caregivers. Mounting evidence suggests that these changes are more than just symptoms. Altered rhythms may contribute to AD progression and development. Many important transcripts, proteins, and metabolites oscillate with a daily cycle. Understanding these rhythms, and their influence on AD, offers the potential to identify new therapies. The translation of circadian biology to Alzheimer's care is limited. Which molecules and pathways show daily rhythms in our human brains? How do those rhythms change with AD? Do changes in molecular rhythms explain changing behavioral patterns? Can these rhythms be exploited for therapeutic benefit? To answer clinical questions, we need human data. AD brain banks provide an invaluable resource. But brain banks almost never provide the time of day when patients died, making it difficult to use these data for rhythms research. We developed CYCLOPS (CYCLic Ordering by Periodic Structure), a machine-learning tool to uncover molecular rhythms using unordered biopsy samples. Evaluating brain expression data, we showed that CYCLOPS could correctly reconstruct rhythms in brain samples and correctly predict the time of death. Here we will order cortical brain samples from control subjects and patients with AD. We will reconstruct the molecular rhythms in these human brains, identifying differences in AD patients and rhythms in known drug targets and AD disease pathways. We will analyze a subset of samples where time of death is known, comparing each subject's “internal molecular time” with the “time on the clock.” We will test the hypothesis that patients with poorly aligned molecular rhythms are more likely to have circadian behavioral disturbance. We will evaluate a measure of transcriptional rhythm strength, testing if “weaker” rhythms predict behavioral or molecular misalignment. Does AD alter rhythm generation? Does it desynchronize still rhythmic cells and brain regions? Using data from multiple brain regions sampled from the same subjects, we will evaluate intracortical circadian synchrony and compare AD patients with controls. Using single-nucleus sequencing data, we will explore the effect of AD on cell type specific rhythms and their synchrony. Finally, we will test the direct influence of important AD causing mutations on molecular clock function, measuring rhythms in isolated cells. This work will advance our understanding of circadian rhythms in AD pathology, clarify the relationship between behavioral and molecular circadian disruption, and catalyze opportunities for AD chronotherapy.