Skip to main content

Advertisement

Reduction of multi-compartmental biophysical models by incremental, automated retuning of their parameters and synaptic weights

Article metrics

  • 626 Accesses

When simulating large networks of neuronal models with conductance-based membrane dynamics, neuronal models with large numbers of compartments are often not practical due to the computational resources required. In this case, a common approach is to seek a model with fewer compartments that approximates the behaviour of the full compartmental model [1, 2]. While analytical techniques to reduce multi-compartmental models consisting of passive, quasi-active or restricted classes of active models exist [35], in general, the reduction of active models requires parameter retuning in order to retain functionally relevant properties of the model. We propose an incremental, automated parameter retuning approach to reduce the number of compartments, which avoids labor-intensive manual retuning and ensures key functional properties of the model are maintained.

Starting from the full compartmental model, the number of compartments is incrementally reduced by merging compartment subsets into single, electronically equivalent compartments. After each compartment reduction phase, all parameters of the model, including synaptic weights, are locally optimised to best match the dynamics and post-synaptic potentials of the full compartmental model using multiple objectives, including phase-plane trajectory histograms of membrane voltage and internal states [6, 7]. The successive local optimisations during the incremental reduction in the number of compartments ensures the model parameters do not transition to qualitatively distinct regions of the parameter space. The multi-compartmental models are iteratively reduced in this manner until key functional properties can no longer be maintained within specified tolerances.

The proposed algorithm is demonstrated on multi-compartmental models of neurons in the cerebellum.

References

  1. 1.

    Bush PC, Sejnowski TJ: Reduced compartmental models of neocortical pyramidal cells. J. Neurosci. Meth. 1993, 46: 159-166. 10.1016/0165-0270(93)90151-G.

  2. 2.

    Davison AP, Feng J, Brown D: A reduced compartmental model of the mitral cell for use in network models of the olfactory bulb. Brain Res Bull. 2000, 51 (5): 393-9. 10.1016/S0361-9230(99)00256-7.

  3. 3.

    Rall W: Branching dendritic trees and motoneuron membrane resistivity. Exp Neurol. 1959, 1 (5): 491-527. 10.1016/0014-4886(59)90046-9.

  4. 4.

    Ohme M, Schierwagen A: An equivalent cable model for neuronal trees with active membrane. Biol Cybern. 1998, 78 (3): 227-43. 10.1007/s004220050429.

  5. 5.

    Hedrick KR, Cox SJ: Structure-preserving model reduction of passive and quasi-active neurons. J Comput Neurosci. 2013, 34 (1): 1-26. 10.1007/s10827-012-0403-y.

  6. 6.

    Druckmann S, Banitt Y, Gidon A, Schürmann F, Markram H, Segev I: A novel multiple objective optimization framework for constraining conductance-based neuron models by experimental data. FNINS. 2007, 1 (1): 7-18. 10.3389/neuro.01.1.1.001.2007.

  7. 7.

    Van Geit W, Achard P, De Schutter E: Neurofitter: a parameter tuning package for a wide range of electrophysiological neuron models. Frontiers in Neuroinformatics. 2007, 1: 1-18.

Download references

Author information

Correspondence to Thomas G Close.

Rights and permissions

This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Keywords

  • Local Optimisation
  • Computational Resource
  • Internal State
  • Active Model
  • Large Network