Graph-based analysis of volumetric image data reveals predominant layer Va-to-II/III feedback in mouse motor cortex

Area of Science:

• Neuroscience
• Computational Neuroscience
• Systems Neuroscience

Background:

• Precise 3D interactions among cortical neurons are crucial for layer-specific computations but remain poorly understood.
• Understanding neural circuit dynamics is key to deciphering brain function, particularly in motor control.

Purpose of the Study:

• To develop a graph framework for inferring functional connectivity from neural activity.
• To investigate the precise 3D interactions and information flow within the mouse primary motor cortex.

Main Methods:

• Utilized fast volumetric two-photon Ca2+ imaging of spontaneous activity in awake mice.
• Developed a graph framework to reconstruct a directed, weighted network of ~1,000 neurons from deconvolved Ca2+ traces.
• Applied layer-specific thresholds and decomposed the network into strongly connected sub-networks.

Main Results:

• Identified ~30 sub-networks, primarily in layers II/III, often bridging to layer Va.
• Found that layer II/III dominates cortical connectivity.
• Observed that feedback connections (Va → II/III) are more prevalent than feedforward connections (II/III → Va).
• Determined that information flows through a maximum of 6 synapses within these sub-networks.
• Uncovered seven distinct geometrical and dynamical motifs representing column-like microcircuits.

Conclusions:

• Cortical microcircuits in the primary motor cortex exhibit diverse, column-like structures with a net ascending information flow.
• These identified sub-networks likely function as elemental processing modules for motor control.
• The developed graph framework provides a powerful tool for analyzing complex neural circuit dynamics.