Non-linear dimensionality reduction on extracellular waveforms reveals cell type diversity in premotor cortex.

Area of Science:

• Neuroscience
• Computational Neuroscience
• Systems Neuroscience

Background:

• Cortical circuits comprise numerous cell types crucial for behavior.
• Existing methods for identifying cell types from extracellular waveforms capture limited variation.

Purpose of the Study:

• To introduce WaveMAP, a new computational method for identifying putative cell types in cortical circuits.
• To apply WaveMAP to extracellular recordings from macaque monkey dorsal premotor cortex.

Main Methods:

• WaveMAP combines non-linear dimensionality reduction with graph clustering.
• Analysis of extracellular waveforms from macaque monkeys during a decision-making task.

Main Results:

• WaveMAP robustly identified eight distinct waveform clusters.
• These clusters confirmed known cell types and revealed novel subtypes.
• Identified cell types showed unique laminar distributions, firing patterns, and decision-related dynamics.

Conclusions:

• WaveMAP provides a more nuanced understanding of cortical cell type diversity and dynamics.
• This method surpasses traditional feature-based approaches in uncovering cellular heterogeneity.