CNS Workshop 2008

Interoperability of Software for Computational and Experimental Neuroscience

Organizers: Padraig Gleeson and Sharon Crook

Abstract: Biophysically detailed computational models are increasingly accepted as important tools for the investigation of brain function by the wider neuroscience community. However, there are still a number of issues to address before a clear and practical framework can be created for exchange of ideas and data both between theoreticians working in different areas and between modelers and experimentalists. At present, multiple simulation platforms are used to model cellular and network activity, each of which has its own scripting language and data structures. This can make reuse of model code developed for one environment difficult for users of another platform, despite the fact that the physiological concepts underlying software design are the same in both. Also, software applications for analysis and management of data produced by electrophysiological experiments, and tools and utilities for the analysis of simulation results are normally developed independently, although the same analysis techniques can be carried out on both datasets.

This workshop includes presentations from researchers who are actively involved in the construction of software solutions for various stages of the computational modeling cycle: from obtaining experimental results, to model creation, simulation and analysis, to prediction of experimental results, and back again. The aim is to present an overview of initiatives in the field to allow greater interaction between these elements and increased usability of results from each stage.

After the main talks, an open discussion session will a) identify "gaps" in the tool chain and b) identify desired extensions/updates to existing standards that allow for greater biophysical detail in models.

Schedule:

9:00	Sharon Crook, Arizona State University Introduction to workshop
9:15	Pierre Yger, Centre National de la Recherche Scientifique PyNN: A common Python interface for neuronal network simulators
9:45	Padraig Gleeson, University College London Enabling interoperability and transparency of models of biophysical neurons and networks with NeuroML
10:15	Hugo Cornelis, University of Texas Health Science Center at San Antonio A technical overview of the CBI simulation framework: Examples of instances and applications
10:45	Coffee Break
11:15	Cengiz Gunay, Emory University Standardizing acquired and simulated electrophysiological data using the hierarchical data format (HDF5)
11:45	Subhasis Ray, National Centre for Biological Sciences, Bangalore MOOSE: The Multiscale Object-oriented Simulation Environment
12:15	Lunch Break
14:00	Ivan Raikov, Okinawa Institute of Science and Technology Neuroscience modelling languages: Practice and theory
14:30	Dan Goodman, Ecole Normale Superieure, Paris Brian: A simulator for spiking neural networks in Python
15:00	Phillip Lord, Newcastle University The CARMEN Project: Towards a common data format for electrophysiological data exchange and analysis
15:30	Break
16:00	Open Discussion

Topics for Open Discussion:

Identification of gaps in the toolchain
Procedural vs. declarative issues
Are standards possible for simulator and experimental data?
Incorporation of kinetic schemes and Markov model descriptions of channels and synapses
Interaction of NeuroML and the computational neuroscience domain with SBML and MathML
Are there issues that we are not considering that provide obstacles to model and data exchange?