Debian Science Modeling of neural systems packages

NEURON est un environnement de simulation pour modéliser des neurones individuels et des réseaux de neurones. Il fournit des outils pour construire, gérer et utiliser commodément des modèles, d'une manière qui ressemble à des nombres et qui soit efficace du point de vue du calcul. C'est particulièrement adapté pour des problèmes fortement liés à des données expérimentales, en particulier ceux impliquant des cellules aux propriétés anatomiques et biophysiques complexes.

NEURON offre

 –⋅une « syntaxe naturelle », qui permet de spécifier les
   propriétés d'un modèle dans des langages courants ;
 –⋅une discrétisation spatiale et temporelle efficace et facile ;
 –⋅plusieurs méthodes d'intégration numérique au choix de
   l'utilisateur ;
 –⋅une interface utilisateur commode (interpréteur et
   interface graphique) ;
 –⋅une bibliothèque, extensible par l'utilisateur, de mécanismes
   biophysiques.

Brian is a clock-driven simulator for spiking neural networks. It is designed with an emphasis on flexibility and extensibility, for rapid development and refinement of neural models. Neuron models are specified by sets of user-specified differential equations, threshold conditions and reset conditions (given as strings). The focus is primarily on networks of single compartment neuron models (e.g. leaky integrate-and-fire or Hodgkin-Huxley type neurons). Features include:

• a system for specifying quantities with physical dimensions
• exact numerical integration for linear differential equations
• Euler, Runge-Kutta and exponential Euler integration for nonlinear differential equations
• synaptic connections with delays
• short-term and long-term plasticity (spike-timing dependent plasticity)
• a library of standard model components, including integrate-and-fire equations, synapses and ionic currents
• a toolbox for automatically fitting spiking neuron models to electrophysiological recordings

PyNN allows for coding a model once and run it without modification on any simulator that PyNN supports (currently NEURON, NEST, PCSIM and Brian). PyNN translates standard cell-model names and parameter names into simulator-specific names.

XPPAUT is a tool for solving

• differential equations,
• difference equations,
• delay equations,
• functional equations,
• boundary value problems, and
• stochastic equations.

The code brings together a number of useful algorithms and is extremely portable. All the graphics and interface are written completely in Xlib which explains the somewhat idiosyncratic and primitive widgets interface.

NEST is a simulation system for large networks of biologically realistic point-neurons and neurons with a small number of electrical compartments.

PCSIM is a tool for simulating heterogeneous networks composed of different model neurons and synapses. This simulator is written in C++ with a primary interface to the programming language Python. It is intended to simulate networks containing up to millions of neurons and on the order of billions of synapses. This is achieved by distributing the network over different nodes of a computing cluster by using MPI.

The iLab Neuromorphic Vision C++ Toolkit (iNVT, pronounced ``invent'') is a comprehensive set of C++ classes for the development of neuromorphic models of vision. Neuromorphic models are computational neuroscience algorithms whose architecture and function is closely inspired from biological brains. The iLab Neuromorphic Vision C++ Toolkit comprises not only base classes for images, neurons, and brain areas, but also fully-developed models such as our model of bottom-up visual attention and of Bayesian surprise.

Features at a glance:

• Low-level neural network simulation classes.
• High-level neuromorphic classes.
• Neuromorphic models of visual attention.
• Hardware interfacing
• Parallel processing classes for the simulation of complex models.
• Neuromorphic modeling environment.

MOOSE is the Multiscale Object-Oriented Simulation Environment. It is the base and numerical core for large, detailed simulations including Computational Neuroscience and Systems Biology. MOOSE spans the range from single molecules to sub-cellular networks, from single cells to neuronal networks, and to still larger systems. It is backwards-compatible with GENESIS, and forward compatible with Python and XML-based model definition standards like SBML and MorphML. MOOSE is coordinating with the GENESIS-3 project towards the goals of developing educational resources for modeling.