Modeling functional cell types in spike train data.

Area of Science:

• Computational Neuroscience
• Systems Neuroscience
• Neuroscience

Background:

• Accurate computational models of neuronal activity are crucial for understanding brain function.
• Current models often lack the precision to capture the diversity of neuronal roles within circuits.
• Interpreting neuronal function requires refined models that account for cell-specific properties.

Approach:

• Developed a hierarchical generative model for functional cell types, single-cell parameters, and neural responses.
• Derived an expectation-maximization algorithm with variational inference to optimize model likelihood.
• Applied a simultaneous method to estimate cell types and fit single-cell models from simulated and in vitro neural recordings.

Key Points:

• The method accurately recovers ground truth parameters from simulated data.
• It improves the prediction of single-cell activity in mouse primary visual cortex recordings.
• Discovered cell-type clusters are well-separated, generalizable, and interpretable.

Conclusions:

• Explicitly modeling shared functional properties across neurons refines neural response models.
• This approach offers a powerful tool for dissecting neural circuit function.
• Functional cell typing provides a framework for integrating diverse neurobiological data.