Portal de Periódicos da CAPES

Netbuilder: A Tool for Constructing Large and Complex Neural Networks

2009; Frontiers Media; Volume: 3; Linguagem: Inglês

10.3389/conf.neuro.01.2009.11.068

1662-4548

De-Castro-Aguiar Paulo,

EEG and Brain-Computer Interfaces

Event Abstract Back to Event Netbuilder: A Tool for Constructing Large and Complex Neural Networks Maria Mafalda-Sousa1 and Paulo De-Castro-Aguiar1* 1 Universidade do Porto, Instituto de Biologia Molecular Unidade de Morfofisiologia do Sistema Somatosensitivo, Portugal Building large networks with complex architectures can be quite challenging especially if the network is functionally heterogeneous. Here we present NetBuilder, a tool for constructing neural networks with a graphical interface, which solves the presented problem. It specifically allows the user to define the architecture, dynamics, properties and states of the neural network. It is extremely easy to use and very versatile, making the process of building or editing large and complex networks very simple. Network model information entry in NetBuilder relies in connectivity tables. After establishing the number of neural populations present in the model, as well as some global parameters, it is possible to specify particular synaptic properties for every possible populations pair, on top of a connectivity table. This way, each fiber pathway can be modeled individually using the available properties and the dynamics of the synaptic conductances and plasticity. These dynamic models can be edited or new user-defined models can be added to external .lib files. All connections between pre-synaptic as post-synaptic populations are random and the sources choice follows a normal distribution, which defines a topographic receptive field. NetBuilder stores the network model data in two types of files: .dat files, which store all NetBuilder inserted data and .net files which convert all data into a network file which is prepared to be loaded and simulated in NeuralSyns simulation engine (http://sourceforge.net/projects/neuralsyns/). The main proposal of this new version of Netbuilder is to make it available as a free and open source application. It was written in C++ programming language, using a cross- platform library to construct the graphical interface. NetBuilder was created on Linux systems but it is prepared to be compiled and used regardless of the platforms (Windows, Linux, MacOS). New features were included in order to facilitate the user interaction with the interface and to optimize the computational performance in the network creation. Unlike several applications of this nature, as Neuron, Genesis, NEST or CSIM, NetBuilder allows you to build or edit large and complex network models without writing a single line of code. Furthermore, if changes have to be subsequently applied to the model, it provides a flexible interface to edit specific parameters and produce the appropriate updated network model.Supported by FCT Grant. Conference: 11th Meeting of the Portuguese Society for Neuroscience, Braga, Portugal, 4 Jun - 6 Jun, 2009. Presentation Type: Poster Presentation Citation: Mafalda-Sousa M and De-Castro-Aguiar P (2009). Netbuilder: A Tool for Constructing Large and Complex Neural Networks. Front. Neurosci. Conference Abstract: 11th Meeting of the Portuguese Society for Neuroscience. doi: 10.3389/conf.neuro.01.2009.11.068 Copyright: The abstracts in this collection have not been subject to any Frontiers peer review or checks, and are not endorsed by Frontiers. They are made available through the Frontiers publishing platform as a service to conference organizers and presenters. The copyright in the individual abstracts is owned by the author of each abstract or his/her employer unless otherwise stated. Each abstract, as well as the collection of abstracts, are published under a Creative Commons CC-BY 4.0 (attribution) licence (https://creativecommons.org/licenses/by/4.0/) and may thus be reproduced, translated, adapted and be the subject of derivative works provided the authors and Frontiers are attributed. For Frontiers’ terms and conditions please see https://www.frontiersin.org/legal/terms-and-conditions. Received: 07 Aug 2009; Published Online: 07 Aug 2009. * Correspondence: Paulo De-Castro-Aguiar, Universidade do Porto, Instituto de Biologia Molecular Unidade de Morfofisiologia do Sistema Somatosensitivo, Porto, Portugal, pauloaguiar@fc.up.pt Login Required This action requires you to be registered with Frontiers and logged in. To register or login click here. Abstract Info Abstract The Authors in Frontiers Maria Mafalda-Sousa Paulo De-Castro-Aguiar Google Maria Mafalda-Sousa Paulo De-Castro-Aguiar Google Scholar Maria Mafalda-Sousa Paulo De-Castro-Aguiar PubMed Maria Mafalda-Sousa Paulo De-Castro-Aguiar Related Article in Frontiers Google Scholar PubMed Abstract Close Back to top Javascript is disabled. Please enable Javascript in your browser settings in order to see all the content on this page.