Connectomes: a re-emerging hub dataset for neuroscience

Abstract

Connectomics aims to generate complete neural wiring diagrams and leverage them to explain brain function at a systems level. Traditional neuroscience has relied on sparse recordings and partial circuit reconstructions, limiting mechanistic insight into how distributed neural activity produces behavior. By contrast, connectomes provide comprehensive maps of all neurons and all synapses, analogous in some respects to the transformative impact of whole-genome sequencing in biology. Early efforts, including the Caenorhabditis elegans connectome, required extensive manual reconstruction, but recent advances in electron microscopy, light-based imaging, and artificial intelligence now enable large-scale datasets such as the full nervous system connectomes of Drosophila melanogaster. These datasets are increasingly integrated into experimental and computational research to define neuronal cell types, reconstruct fine-scale circuits, and analyze network organization. Connectomics has enabled the discovery of novel neuron types, clarified sensory and motor pathways, and facilitated predictive models of neural computation. Beyond anatomical reconstruction, connectome graphs increasingly serve as substrates for network analysis, simulation, and perturbation, providing a framework for generating and testing mechanistic hypotheses about brain function. Network studies reveal conserved principles, including small-world topology and rich-club organization, while connectome-constrained simulations and perturbation approaches offer new opportunities for linking structure to function. Despite these advances, connectomes remain limited by small sample sizes, scale, cost, and their fundamentally static nature, lacking direct information on neural dynamics, physiological state, and plasticity. Integrating structural data with functional, neuromodulatory, glial, and behavioral measurements remains a major challenge. Nonetheless, connectomes have emerged as foundational hub datasets that organize diverse streams of neuroscientific evidence and provide an increasingly important framework for advancing a mechanistic understanding of brain function and dysfunction.