Python in neuroscience

• Published in: Frontiers in neuroinformatics Vol. 9; p. 11
• Main Authors: Muller, Eilif, Bednar, James A., Diesmann, Markus, Gewaltig, Marc-Oliver, Hines, Michael, Davison, Andrew P.
• Format: Journal Article
• Language: English
• Published: Switzerland Frontiers Research Foundation 14.04.2015; Frontiers Media; Frontiers Media S.A
• Subjects: Alzheimer's disease; Artificial intelligence; Basal ganglia; Bioinformatics; Cognitive Sciences; collaboration; Data analysis; Data processing; Firing pattern; Hippocampus; Hypothesis testing; Information theory; Interfaces; Interoperability; Libraries; Life Sciences; Machine learning; Medical imaging; Navigation behavior; Nervous system; Neural coding; Neurobiology; Neurons and Cognition; Neuroscience; Neurosciences; Open source software; Programming languages; Psychology and behavior; python language; Researchers; scientific computing; Signal transduction; Software Development; Aachen Germany; Germany; python language; interoperability; software development; scientific computing; collaboration
• ISSN: 1662-5196; 1662-5196
• DOI: 10.3389/fninf.2015.00011

(2009) describe the use of Python for information-theoretic analysis of neuroscience data, outlining algorithmic, statistical and numerical challenges in the application of information theory in neuroscience, and explaining how the use of Python has significantly improved the speed and domain of applicability of the algorithms, allowing more ambitious analyses of more complex data sets. [...]in the domain of electrophysiology, Garcia and Fourcaud-Trocmé (2009) describe OpenElectrophy, an application for efficient storage and analysis of large electrophysiology datasets, which includes a graphical user interface for interactive visualization and exploration and a library of analysis routines, including several spike-sorting methods. The range of modeling domains of these simulators is wide, from stochastic simulation of coupled reaction-diffusion systems (STEPS), through simulation of morphologically detailed neurons and networks (NEURON, MOOSE), highly-efficient large-scale networks of spiking point neurons (NEST, PCSIM, NCS, Brian) to population coding or point-neuron models of large brain regions (Nengo, Topographica). The addition of Python interfaces to such a large number of widely used simulation environments suggested a huge opportunity to enhance interoperability between different simulators, making use of the common scripting language, which in turn has the potential to enhance the transfer of technology, knowledge and models between users of the different simulators, and to promote model reuse.