Comparing realistic subthalamic nucleus neuron models

Felix Njap*, Jens C. Claussen, Andreas Moser, Ulrich G. Hofmann

*Corresponding author for this work

Research output: Chapter in Book/Published conference outputConference publication


The mechanism of action of clinically effective electrical high frequency stimulation is still under debate. However, recent evidence points at the specific activation of GABA-ergic ion channels. Using a computational approach, we analyze temporal properties of the spike trains emitted by biologically realistic neurons of the subthalamic nucleus (STN) as a function of GABA-ergic synaptic input conductances. Our contribution is based on a model proposed by Rubin and Terman and exhibits a wide variety of different firing patterns, silent, low spiking, moderate spiking and intense spiking activity. We observed that most of the cells in our network turn to silent mode when we increase the GABAA input conductance above the threshold of 3.75 mS/cm2. On the other hand, insignificant changes in firing activity are observed when the input conductance is low or close to zero. We thus reproduce Rubin's model with vanishing synaptic conductances. To quantitatively compare spike trains from the original model with the modified model at different conductance levels, we apply four different (dis)similarity measures between them. We observe that Mahalanobis distance, Victor-Purpura metric, and Interspike Interval distribution are sensitive to different firing regimes, whereas Mutual Information seems undiscriminative for these functional changes.

Original languageEnglish
Title of host publication2011 International Symposium on Computational Models for Life Sciences, CMLS-11
Number of pages8
ISBN (Print)9780735409316
Publication statusPublished - 13 Jul 2011
Event2011 International Symposium on Computational Models for Life Sciences, CMLS-11 - Toyama City, Japan
Duration: 11 Oct 201113 Oct 2011

Publication series

Name AIP Conference Proceedings
ISSN (Print)0094-243X


Conference2011 International Symposium on Computational Models for Life Sciences, CMLS-11
CityToyama City

Bibliographical note

© AIP 2011. Comparing Realistic Subthalamic Nucleus Neuron Models
AIP Conference Proceedings 1371, 102 (2011);
Felix Njap, Jens C. Claussen, Andreas Moser, and Ulrich G. Hofmann


  • γ-aminobutyric acid (GABAA) receptors
  • Computational model
  • Similarity measures
  • Spike-trains


Dive into the research topics of 'Comparing realistic subthalamic nucleus neuron models'. Together they form a unique fingerprint.

Cite this