DSSS - Bonhoeffer memorial lecture: Structural neurobiology in zebrafish

Higher brain functions arise from the exchange of information between neurons in large networks. The function of such networks is determined to a large extent by their “wiring diagrams”, i.e. the specific synaptic connectivity between individual neurons. Moreover, information is assumed to be stored in neuronal networks by specific modifications of synaptic wiring diagrams, both during evolution and over the lifetime of individuals. The anatomical reconstruction of neuronal wiring diagrams has, however, been a major challenge because it requires ultrastructural resolution (to identify synapses) throughout large volumes (to reconstruct populations of neurons). Recently, pioneering developments in volume electron microscopy paved the way for the reconstruction of neuronal circuits consisting of thousands of neurons at synaptic resolution. In an approach referred to as “dynamical connectomics”, we combine the dense reconstruction of neuronal circuits with behavioral analyses and large-scale optical measurements of neuronal population activity in zebrafish. Dynamical connectomics revealed a mechanism underlying transformations of distributed neuronal activity patterns in the olfactory system that facilitate the discrimination of similar sensory inputs by associative memory networks. This work demonstrated that detailed analyses of wiring diagrams, together with measurements of neuronal activity and computational modeling, can provide mechanistic insights into complex neuronal computations that require higher-order network structure. I will present an overview of this and other work that uses dynamical connectomics to understand neuronal computations underlying higher brain functions. Moreover, I may present results from studies using activity measurements during behavior of zebrafish in a virtual reality to examine how neuronal populations predict future sensory inputs.