Ultrastructural mapping of neural circuitry: A computational framework

Abstract

Complete mapping of neuronal networks requires data acquisition at synaptic resolution with canonical coverage of tissues and robust neuronal classification. Transmission electron microscopy (TEM) remains the optimal tool for network mapping. However, capturing high resolution, large, serial section TEM (ssTEM) image volumes is complicated by the need to precisely mosaic distorted image tiles and subsequently register distorted mosaics. Moreover, most cell or tissue class markers are not optimized for TEM imaging. We present a complete framework for neuronal reconstruction at ultrastructural resolution, allowing the elucidation of complete neuronal circuits. This workflow combines TEM-compliant small molecule profiling with automated image tile mosaicking, automated slice-to-slice image registration and terabyte-scale image browsing for volume annotation. Networks that previously would require decades of assembly can now be completed in months, enabling large-scale connectivity analyses of both new and legacy data. Additionally, these approaches can be extended to other tissue or biological network systems.