Project Summary The robust, yet sensitive, systems underlying daily behaviors provide an outstanding opportunity to apply dynamical network science to understand how neural connectivity adapts to a changing environment. Daily rhythms in behavior and physiology (e.g., sleep-wake and hormone secretion) depend on circadian clock genes, pacemaking cells and cell-cell signaling to adjust to daily cues such as seasonal changes in daylength. The suprachiasmatic nucleus, SCN, coordinates these daily rhythms to anticipate challenges such as finding food and mates and avoiding predators. This application focuses on major gaps in our understanding of how SCN cells adapt and synchronize to produce daily rhythms in response to seasonal changes in photoperiod. The proposed experiments combine in vivo and in vitro cell-type specific perturbations and recordings with computational biology and control engineering to test the central hypothesis that specific and reversible changes in network topology underlie adaptation to long and short days. The Aims will map, for the first time, cell-cell connectivity changes in the SCN during adjustment to long and short days. We will record gene expression and intracellular calcium from distinct classes of neurons within the SCN before, during, and after exposure to long (summer) or short (winter) days and, using a novel data science method, infer their connectivity. We will evaluate these inferred networks by using computational models to predict their behavior to control- theory-inspired perturbations that are then implemented on identified cells in vitro and in vivo. Taken together, the proposed aims will elevate the SCN into the small class of circuits that has been mapped with sufficient cellular resolution to allow evaluation of connectivity rules that support the robust, yet sensitive, network performance. The results from the experiments will lay a foundation for understanding how the brain is organized as a network of synchronized circadian cells. We will create methods to reveal functional neural topology and to analyze and control dynamic structures in complex networks with different spatial and temporal scales. Ultimately, these experiments will provide guidelines to reveal, and insights to understand, how diverse network topologies adapt to be robust and yet sensitive to everyday cues.