Epilepsy is the third most prevalent neurological disorder after stroke and Alzheimer’s Disease with an incidence of 1 in 26 individuals. It is estimated that 3 million people in the U.S and 65 million worldwide currently live with epilepsy. Approximately 30% of epilepsy patients are drug resistant and this number has remained the same since 1850 despite the marketing of 12 new antiseizure drugs in the past 20 years. A major hurdle to understanding the disease is that seizures are transient and importantly, difficult to predict. This prevents the acquisition of a detailed portrait of molecular, cellular and circuit alterations at the most critical time: just before seizure onset. However, seizures manifest a clear temporal organization: they display circadian and multi day (mulitdien) rhythmicity in patients, regardless of epilepsy type or affected brain region. Combining circadian and multidien rhythms, one can extract high seizure risk (HiSR) and low seizure risk (LoSR) epochs in a patient- specific manner. Importantly, this cyclicity in risk exists in two rat models of epilepsy and canine epilepsy. The existence of HiSR and LoSR times imply that circuit excitability changes in a periodic manner that can be modelled and predicted, and its mechanisms studied. We have developed a machine learning tool that acquires continuous EEG over many weeks, learns the pattern of interictal activity and then predicts when an animal is entering a LoSR or HiSR epoch. The goal of this application is to embark on the first ever global molecular and cellular exploration of HiSR and LoSR epochs. Historically, a common research strategy has been to compare non-epileptic to epileptic brains and, clearly this approach has yielded a wealth of knowledge regarding mechanisms behind seizure genesis and epileptogenesis. The premise of our application is that epileptic and non-epileptic brains are different enough from each other that additional, unique insights will be uncovered by using LoSR epochs as controls for HiSR epochs. Preliminary data from bulk tissue shows a differential expression of nearly 100 hippocampal proteins between HiSR and LoSR times. Together, these findings inform our central hypothesis that large scale hippocampal gene changes, driven by a few Master regulators, contribute to alterations in seizure risk over the multidien cycle. We will perform single cell RNAseq on hippocampi from rats in LoSR and HiSR to generate a high resolution, single cell map of all gene changes that occur as a brain transitions from a low to high seizure risk state. To the best of our knowledge this will be the first ever study comparing molecular and electrophysiological changes at the single cell level in the epileptic brain as it transitions from a low to a high seizure risk state. Using bioinformatic tools developed and published by us, we will identify potential Master Regulators behind these changes in every cell type and subfield of the rat hippocampus. The major del...