Automated computation of arbor densities: a step toward identifying neuronal cell types

Area of Science:

• Neuroscience
• Computational Biology
• Cell Biology

Background:

• Neuronal structure, including shape and position, is key to identifying cell types and understanding function.
• Traditional methods like arbor tracing are time-consuming, hindering large-scale neural circuit analysis.

Purpose of the Study:

• To develop an automated method for quantifying dendritic arbor density in neurons.
• To establish that mouse retinal ganglion cell types precisely distribute arbor volumes within the inner plexiform layer.

Main Methods:

• Three-dimensional reconstruction of neuronal arbors using supervised machine learning.
• Post-processing to remove cell bodies and isolate individual neurons.
• Registration of neurons using starburst amacrine interneurons as fiducial markers.

Main Results:

• Demonstrated precise distribution of arbor volumes across the inner plexiform layer by mouse retinal ganglion cell types.
• Developed and validated an automated approach for computing arbor density relative to depth.

Conclusions:

• Automated arbor density mapping offers a faster alternative to manual tracing for neuronal analysis.
• The developed method shows potential for generalization to other central nervous system structures with sparse labeling and spatial reference.