Computational Classification Approach to Profile Neuron Subtypes from Brain Activity Mapping Data.

Area of Science:

• Neuroscience
• Computational Biology
• Systems Neuroscience

Background:

• Understanding cell type-specific neural activity is crucial for deciphering how neural circuits generate cognition.
• Existing in vivo neurophysiological datasets lack comprehensive exploration of these activity patterns.

Purpose of the Study:

• To introduce a novel computational approach, inter-spike-interval classification-analysis (ISICA), for uncovering distinct cell subpopulations from in vivo neural spike datasets.
• To demonstrate the utility of ISICA in classifying neuron subtypes and linking them to specific physiological functions.

Main Methods:

• ISICA involves four steps: spike pattern feature extraction, pre-clustering analysis, clustering classification, and unbiased classification-dimensionality selection.
• The method utilizes gamma distribution shape factors and coefficient of variation of inter-spike intervals to analyze spike dynamics.

Main Results:

• ISICA achieved invariant classification of dopaminergic neurons and CA1 pyramidal cell subtypes across different brain states.
• Classified dopaminergic neuron subtypes differentially encoded aspects of fear, such as valence and value.
• Distinct hippocampal CA1 pyramidal cell subtypes showed differential responses to ketamine-induced anesthesia.

Conclusions:

• The ISICA method provides a robust tool for data mining large-scale in vivo neural datasets.
• This approach facilitates novel insights into neural circuit dynamics underlying cognition.
• ISICA enables the identification of functionally distinct neuron subtypes crucial for understanding brain states and behaviors.