Detailed Modelling of
Signal Processing in Neurons
Personal tools


The analysis of information processing neurons concentrates on how signaling systems are used in the adult brain to respond to environmental cues. The utmost challenge in the analysis of complex signaling systems is to obtain a mechanistic understanding of signal processing in neurons based on subcellular resolution. Although numerous neuronal processes have been studied in great detail and the information available on the biochemistry and genetics of neurons is increasing rapidly, still central questions remain open. A central goal of the Heidelberg neuroscience community is to pursue a multi-disciplinary approach that translates experimental work into a mathematical language and describes in realistic computational models how arrays of signaling systems act and interact within and between neurons. Starting at the molecular level, regulatory modules of increasing complexity are being characterized using conceptually and methodologically similar frameworks. The interrelationship of individual units and processes will be investigated using numerical simulation on realistic geometries with full 3-dimensional resolution as a key tool for a quantitative understanding of the relevant processes. To foster the exchange between modellers and the experimentalists, an additional permanent junior position in computational neuroscience will be installed, to play a key role in bridging the gap between theory and application. Ultimately, we hope to understand how functional organizations of neurons change as information flows through the system, which may help explain neuronal plasticity.

We address these questions with a hierarchical approach starting from spatially resolved models of calcium signaling, including detailed mathematical description of the most relevant physico-chemical processes up to microcircuits, consisting of few connected neurons, as a paradigm for small local networks. We expect that detailed simulations will lead to a deeper understanding of the dynamics of neurons and microcircuits.

To achieve these goals, creating simulation models of single neurons, of single synapses and of microcircuits is one of the essential requirements. To that end, a hierarchy of simulation models will be set up on several modeling scales. Those models will interact with each other, in order to enable setting up a full three-dimensional volume model of a microcircuit. Starting from models of subcellular processes like Ca2+ signaling and modeling synapses, fully coupled simulation models have to be set up and run on the full geometry of a microcircuit. The models shall take into account the spatial heterogeneity, coupling between signal transmission and transport of the main triggering ions Ca2+, Na+, K+, Cl–, the corresponding ion channels and the reaction of calcium with calcium buffers and other linkage partners as well as diffusion and transport in 3 dimensions. Intracellular diffusion of those ions has to be included, since, for example, the concentration of Cl– is strongly influenced by and influences the activity of inhibitory synapses and regulation of transporting molecules.

This implies the following sub goals.

• Development of a valid model of signal transmission in a single cell taking into account the inhomogeneous distribution and the transport of the important ions inside and outside the cell as well as the properties of the membrane.

• Modeling the geometry of a single cell. To that end the tool NeuRA developed in co-operation by MPImF and SiT is a reasonable basis. In addition, tools have to be created to automatically reconstruct the cell surface.

• Generating sensible computing grids resolving the highly complex geometry.

• Detailed modeling of Ca2+-signaling in a single cell and effects on long-term plasticity.

• Modeling and simulation of the ionic conductance of membranes and synapses.

• Detailed modeling of microcircuits coupled by synapses and gap junctions.

• Studying the dynamics of microcircuits by detailed simulation.

• Development of robust simulation methods and tools for solving the coupled system of partial differential equations modeling signal transduction efficiently. In most cases, the simulation framework ≈, developed by SiT, will be the basis for the simulation tool. In view of the complexity to be described this is a task reaching to the limits of the largest computers available nowadays.

• Comparison with compartmental models. As basis for compartmental models with very fine resolution we may the tools NeuGen and NeuSim developed by SiT.

• Validating the simulation tools based on measured data. To that end, a sensitivity analysis and inverse modeling is necessary. This also requires quite a lot of experimental data and a strong interaction between experimental groups and modeling ones.

• Modeling of the Drosophila Larva neuromuscular junction. The synapses simplicity, accessibility to various electrophysiological recording and imaging techniques, and the genetic malleability which is intrinsic to the Drosophila system make it ideal for computational and mathematical studies.

So far no comparable tools exist. We expect that the model to be developed here will bring substantial progress in understanding neuron dynamics.


Document Actions