Navigation and service

Unified framework for spiking and gap-junction interactions

In large scale network simulators like NEST, a common optimization strategy is to communicate spike information at relative long time intervals, limited by the shortest synaptic transmission delay in the network. This approach is at odds with the instantaneous neuronal interactions required for modelling gap junctions (electrical synapses with diverse functional roles in network dynamics). The interaction of the membrane potentials of two neurons across a gap junction is typically represented in the form Igap, ij(t) =  Gij(Vi(t) −  Vj(t)). Simulation of a network of gap junction coupled neurons conceptually amounts to finding the solution of a set of coupled ordinary differential equations, which can be solved using the waveform relaxation method.

In the unified framework project, the Jacobi waveform relaxation method is used to solve gap junction interactions between neurons in the NEST simulator (Hahne et al., 2015). The implemented algorithm based on this method allows for a trade-off between numerical precision and communication size between neurons in the simulation. The algorithm and required data structures are suitable for distributed computing and deliver high performance and high accuracy.

Gap Junction

Figure: Two communication strategies using the waveform relaxation technique in the NEST simulator. (A) The membrane potentials are communicated in intervals equal to the minimal synaptic delay, increasing the amount of computation needed to reach acceptable accuracy. (B) Potentials are communicated in every computation time step h, increasing the amount of data to be exchanged between nodes of a large scale simulation (from Hahne et al., 2015).

Our Contribution

  • Design and development of the approach
  • Implementation of a prototype
  • Integration into the NEST-simulator

Related publications

Hahne, J., Helias, M., Kunkel, S., Igarashi, J., Bolten, M., Frommer, A., & Diesmann, M. (2015). A unified framework for spiking and gap-junction interactions in distributed neuronal network simulationsFrontiers in Neuroinformatics9, 22.