A New Era in Computational Neuroscience

Author: Blackwell, Kim T.

Source: Neuroinformatics, Volume 3, Number 2, June 2005 , pp. 163-166(4)

Buy & download fulltext article:

Price: $42.00 plus tax (Refund Policy)

Abstract:

Computational neuroscience is about to enter a new era. The original focus of neuronal modeling was to evaluate the influence of ionic channels on neuronal activity, inspired by the work of Hodgkin and Huxley (1952), and to investigate the effect of electrotonic properties on synaptic integration, initiated by Rall (1969). Recently, the complexity of single neuron models has been expanding in parallel with the field of neuroinformatics. The first, and still the most common, increase in complexity of single neuron models is to include morphological detail as revealed by tracing and reconstruction of biocytin filled neurons. Earlier models, e.g., DeSchutter and Bower (1994), tend to use uniform distribution of ionic channels; whereas later models, e.g., Poirozi et al. (2003), sometimes use inhomogeneous channel densities as suggested by electrophysiological findings (Hoffman et al. 1997). The development of such models is facilitated by neuroinformatics databases which compile digitized reconstructions of various neurons (e. g., http://www.krasnow.gmu.edu/L-Neuron/ database/index.html). The objective of these models is not only to evaluate how channel properties and distribution influence spiking activity, but also to form a basis for investigating network connectivity and activity (Scorcioni et al., 2002). The second, and more recent, increase in complexity of single neuron models is to include biochemical reactions and second messenger that interact with ionic channels. The most prevalent second messenger modeled is calcium (cf., Blumenfeld et al., 1992; Holmes and Levy, 1990) due to its effect on calcium dependent potassium currents, and its role in synaptic plasticity. Of increasing importance are models of metabotropic (G protein-coupled) receptors and the downstream kinases and phosphatases directly involved in memory storage and gene regulation (e.g., Bhalla and Iyenger, 1999; Blackwell, 2004). This type of modeling greatly overlaps with systems biology, and benefits from bioinformatics databases such as ExPASy, the proteomics server of the Swiss Institute of Bioinformatics, with links to BRENDA (http:// www.brenda.uni-koeln.de/) which includes enzyme kinetics. The unique aspects of modeling second messenger pathways in neurons is the interaction with ionic channels, and compartmentalization by morphological features such as dendrites and spines.

Document Type: Miscellaneous

Publication date: 2005-06-01

A New Era in Computational Neuroscience

Buy & download fulltext article:

Tools

Key

Text size:

Useful pages