Understanding the neural circuit basis of behaviour is one of the great challenges in biology. Yet no single laboratory can study the many brain regions, connections, and cell types working together o coordinate decision-making. There is thus a disparity between the capacity of individual laboratories and the complexity of problems being addressed. The International Brain Laboratory orchestrates the efforts of 21 experimental and theoretical laboratories to understand a single behavior. By standardizing experimental procedures and data pipelines, we enable the data acquired by different laboratories to be assembled into a large combined dataset. Our mouse decision-making task requires valuation of different choices based on sensory evidence, choice selection based on this valuation, and integration of prior experience to update a model of the world. We seek a comprehensive understanding of the neural mechanisms underlying these functions, at scales ranging from single neurons to local microcircuits to inter-area communication. To do so we will harness powerful electrophysiological, optical, and genetic tools, and exploit the rich set of theories and analytical approaches developed in recent years. This large-scale effort will provide the first end-to-end “standard model” of a complex mammalian behavior, and inspire new ways to collaborate on difficult problems in neuroscience.

Our recent collaborative work has delivered whole brain and nerve cord connectomes (synaptic resolution wiring diagrams) for adult Drosophila – the first for an animal that can walk (or fly) with complex visual, motor and cognitive behaviour. We have also identified over 11,000 cell types within these electron microscopy (EM) connectomes, defining their conserved neural circuit architecture. However, only a few hundred of these cell types have been molecularly characterised as far as the genes they express. Moreover, the precise positioning along neurites of key signalling molecules such as gap junctions, fast acting neurotransmitters and neuropeptides and their receptors is a crucial determinant of circuit function, but this information is largely unknown. We propose to optimise new expansion microscopy methods that use total protein staining to label neuronal ultrastructure with EM-like contrast in samples physically expanded up 20x. This super-resolution approach will allow key signalling proteins to be localised within individual neurons and to obtain the complete morphology of all neurons within a brain. In Drosophila neuronal morphologies can be directly matched across brains so this will allow molecular annotation of the connectome. We will focus on Drosophila but the pipeline will be applicable in other species.