An unsupervised map of excitatory neuron dendritic morphology in the mouse visual cortex

Area of Science:

• Neuroscience
• Computational Neuroscience
• Connectomics

Background:

• Neocortical neuron morphology is crucial for neural circuit function.
• Existing classifications of neuron types are discrete.
• Principles governing morphological diversity are not fully understood.

Purpose of the Study:

• To investigate the organizational principles of excitatory neuron morphology in the mouse visual cortex.
• To challenge traditional discrete classifications of neuron types.
• To identify continuous variations in neuronal structure.

Main Methods:

• Utilized graph-based machine learning on over 30,000 reconstructed excitatory neurons.
• Analyzed data from the MICrONS serial-section electron microscopy volume.
• Developed a low-dimensional morphological "bar code" for neuron characterization.

Main Results:

• Cortical excitatory neuron morphology is best described as a continuum, not discrete types, with exceptions in layers 5 and 6.
• Dendritic arbors in layers 2-3 show decreasing width and tuft size with increasing depth.
• Layer 4 exhibited inter-area differences, with V1 having more atufted neurons than higher visual areas.
• Discovered neurons in V1 avoiding deeper layers with their dendrites.

Conclusions:

• Excitatory neuron morphological diversity is better understood through axes of variation rather than distinct morphological types.
• This continuum model provides a more nuanced view of neural structure-function relationships.
• Findings advance our understanding of cortical organization and circuit development.