Computational Neuroscience at UMB, Ås

Geir Halnes

Postdoctoral fellow.
Dept. of Mathematical Sciences and Technology. Norwegian University of Life Sciences.

Mail	PO Box 5003, N-1432 Ås
Office	TF kvartalet, Drøbakveien 31
Fax	+47-64965401
Phone	+47-64965427
E-mail	geir.halnes at umb.no

Current research projects:

Mechanistic modelling of the lateral geniculate nucleus

The lateral geniculate nucleus (LGN) is a part of thalamus which receives visual signals from retinal ganglion cells and transmits processed information to the visual cortex. Traditionally, it is understood as the main mediator between the retina and the cortex. The processing in LGN involves refinement of the receptive field, and temporal decorrelation of visual input. The function and dynamics of the processing differ between sleepy and awake states. State shifts are mediated by input from the cortex and the brain stem.

The LGN consists mainly of two cell types. The main cell type (~75%) is the excitatory relay cells, which project to cortex. The other cell type (~25%) is the and inhibitory, more local interneurons. The relay cells receive input from the retina, and transfer it to the visual cortex. The GABAergic interneurons are thought to shape this information flow, e.g. by enhancing surround inhibition and controlling the number of visually evoked spikes. The LGN interneuron is particularly interesting because of its synaptic outputs, which are both axonal and dendritic, and its synaptic circuits are involved in rare triadic synapses with thalamocortical cells. It is believed that these triads may perform independent computations which are functionally decoupled from the soma.

We approach a mechanistic description of how LGN controls the visual gateway to cortex. This requires biophysically realistic models of the involved cell types, their interactions and the relevant inputs to the LGN.

Compartmental model of the interneurons in the LGN

In collaboration with an experimental group at the University of Oslo, we have developed a biophysically realistic multi-compartmental model of the LGN interneuron. The model includes a realistic morphology (331 compartments), contains 7 different ion channels, which are distributed in a realistic way over the somatodendritic membrane. The simulation tool NEURON ( http://neuron.duke.edu/) was used for the modelling. In this interplay between modeling and experiments, we deduce the responsible mechanisms behind essential features in the LGN interneuron activity. A simplified version of the LGN-interneuron model will later be used in network models, in order to simulate the LGN at a larger scale.

Publication: Halnes G, Augustinaite S, Heggelund P, Einevoll GT, Migliore M (2011). A Multi-Compartment Model for Interneurons in the Dorsal Lateral Geniculate Nucleus. PLoS Comp. Biol. 7:e1002160: web

ModelDB: download

Distribution of T-type calcium channels in LGN interneurons

Independent experimental studies have arrived at different results regarding the distribution of T-type calcium channels on the dendrites of LGN interneurons. We take a computational approach to investigate the functional consequences of different distributions of T-type calcium channels in interneurons. The aim is to infer the correct distribution by comparing our results to a body of observations in published litterature. This project will be a master's project for Joy-Loi Chepkoech.

Inhhibition of relay cell by LGN interneurons under different conditions

We investigate how LGN interneurons affect the output of relay cells under different input conditions and processing states. We use a previously published models of relay cells, and our own interneuron model. Of particular interest are mechanisms that can be related to state dependent shifts in the mode of operation.

Electrodiffusion in Astrocytes

During periods of high neural activity, the extracellular potassium concentrations may become high. One hypothesized function of astrocytes is that they may take up some of this excess potassium, transport potrassium ions intracellularly, and release potassium in regions where the extracellular concentrations of potassium are lower. The transport involves 1) transmembrane currents, 2) intracellular diffusion of potassium ions (due to concentration gradients), 3) intracellular electrical currents containing potassium ions (due to potential gradients). We develop a simple framework for modelling electrodiffusion processes on the time scale of seconds to minutes. The framework is specifically developed for astrocytes, but could be generalized to other cells.