Modeling of lamprey reticulospinal neurons: multiple distinct parameter sets yield realistic simulations

Area of Science:

• Neuroscience
• Computational Biology
• Biophysics

Background:

• Reticulospinal (RS) neurons in vertebrates, including lampreys, are crucial for initiating locomotion by activating spinal motor networks.
• Understanding the biophysical properties of RS neurons is essential for modeling neural control of movement.

Purpose of the Study:

• To construct a biophysically detailed computer model of lamprey RS neurons.
• To systematically explore the parameter space of this model to identify viable sets that accurately replicate neuronal firing properties.

Main Methods:

• A three-compartment model (dendritic, somatic, axon initial segment) was developed, incorporating passive and various voltage-gated ion channels.
• An initial parameter set was manually adjusted and validated against experimental data.
• A dual-annealing search algorithm and grid searches were employed to explore millions of parameter combinations.

Main Results:

• The default model parameter set generated simulations that favorably matched experimental properties of lamprey RS neurons.
• The dual-annealing search identified 4,302 viable parameter sets, with many achieving ~30% improvement in scoring.
• Grid searches revealed significant correlations between the maximum conductances of specific ion channel pairs.

Conclusions:

• Multiple parameter sets can successfully model the action potential and repetitive firing properties of lamprey RS neurons.
• This study represents the first systematic exploration of parameter space for a detailed biophysical model of lamprey RS neurons.
• The findings provide a robust computational framework for further investigating RS neuron function and neural control of movement.