2
BioMed Central Page 1 of 2 (page number not for citation purposes) BMC Neuroscience Open Access Lecture presentation Describing and exchanging models of neurons and neuronal networks with NeuroML Sharon Crook* 1 , R Angus Silver 2 and Padraig Gleeson 2 Address: 1 Dept. of Mathematics and Statistics, School of Life Sciences, Center for Adaptive Neural Systems, Arizona State University, Tempe, Arizona, 85287, USA and 2 Dept. of Neuroscience, Physiology and Pharmacology, University College London, London, WC1E 6BT, UK Email: Sharon Crook* - [email protected] * Corresponding author The Neural Open Markup Language (NeuroML) project is an international, collaborative initiative to facilitate the exchange of complex neural models, allow for greater transparency and accessibility of models, enhance inter- operability between simulators and other tools and sup- port the development of new software and databases [1- 3]. The increasing enthusiasm in the computational neu- roscience community for standards that allow for greater simulator interoperability and model publication is driv- ing current efforts, which focus on the key objects that need to be exchanged among existing applications and try to anticipate those needed by future applications. Exam- ples of these objects include descriptions of neuronal morphology, ion channels, synaptic mechanisms, and network structure. The process of creating these common specifications encourages discussion among users of inde- pendently developed applications, which leads to succinct descriptions of the essential elements of models. Neu- roML is an Open Source project based on XML, as it pro- vides the transparency, portability and extensibility required in these efforts. The openness of the standards and the encouragement of feedback from the community are some of the guiding principles of the NeuroML initia- tive. The declarative specifications for NeuroML are arranged into levels, with higher levels adding extra concepts at dif- ferent spatial scales, an approach that ensures that the specification is provided in a modular way. Mappings exist between NeuroML elements and several commonly used simulators including NEURON [4], GENESIS [5] and PSICS [6], and a number of tools are available which allow a user to create and validate NeuroML documents and to generate code for model implementation by mul- tiple simulators from these documents. In particular, the model development application neuroConstruct can import and write NeuroML documents as well as generate output for simulating neuron or neuronal network activ- ity using either NEURON, GENESIS, PSICS or PyNN [7]. Currently, NEURON can import and export cells in Neu- roML format, and import/export of NeuroML is in beta testing for PyNN, which is a Python package for simulator independent specification of neuronal network models [8]. The use of NeuroML with PyNN provides a connec- tion between the NeuroML descriptions of large-scale neuronal network models and additional simulators. Overall, NeuroML provides a valuable contribution towards simulator interoperability as well as model pub- lication and exchange. The NeuroML standards will facil- itate a broad range of research goals in computational neuroscience. Acknowledgements Partial support for this work was provided by the Wellcome Trust. We also thank the International Neuroinformatics Coordinating Facility and the from Eighteenth Annual Computational Neuroscience Meeting: CNS*2009 Berlin, Germany. 18–23 July 2009 Published: 13 July 2009 BMC Neuroscience 2009, 10(Suppl 1):L1 doi:10.1186/1471-2202-10-S1-L1 <supplement> <title> <p>Eighteenth Annual Computational Neuroscience Meeting: CNS*2009</p> </title> <editor>Don H Johnson</editor> <note>Meeting abstracts – A single PDF containing all abstracts in this Supplement is available <a href="http://www.biomedcentral.com/content/files/pdf/1471-2202-10-S1-full.pdf">here</a>.</note> <url>http://www.biomedcentral.com/content/pdf/1471-2202-10-S1-info.pdf</url> </supplement> This abstract is available from: http://www.biomedcentral.com/1471-2202/10/S1/L1 © 2009 Crook et al; licensee BioMed Central Ltd.

Describing and exchanging models of neurons and neuronal networks with NeuroML

Embed Size (px)

Citation preview

BioMed CentralBMC Neuroscience

ss

Open AcceLecture presentationDescribing and exchanging models of neurons and neuronal networks with NeuroMLSharon Crook*1, R Angus Silver2 and Padraig Gleeson2

Address: 1Dept. of Mathematics and Statistics, School of Life Sciences, Center for Adaptive Neural Systems, Arizona State University, Tempe, Arizona, 85287, USA and 2Dept. of Neuroscience, Physiology and Pharmacology, University College London, London, WC1E 6BT, UK

Email: Sharon Crook* - [email protected]

* Corresponding author

The Neural Open Markup Language (NeuroML) project isan international, collaborative initiative to facilitate theexchange of complex neural models, allow for greatertransparency and accessibility of models, enhance inter-operability between simulators and other tools and sup-port the development of new software and databases [1-3]. The increasing enthusiasm in the computational neu-roscience community for standards that allow for greatersimulator interoperability and model publication is driv-ing current efforts, which focus on the key objects thatneed to be exchanged among existing applications and tryto anticipate those needed by future applications. Exam-ples of these objects include descriptions of neuronalmorphology, ion channels, synaptic mechanisms, andnetwork structure. The process of creating these commonspecifications encourages discussion among users of inde-pendently developed applications, which leads to succinctdescriptions of the essential elements of models. Neu-roML is an Open Source project based on XML, as it pro-vides the transparency, portability and extensibilityrequired in these efforts. The openness of the standardsand the encouragement of feedback from the communityare some of the guiding principles of the NeuroML initia-tive.

The declarative specifications for NeuroML are arrangedinto levels, with higher levels adding extra concepts at dif-ferent spatial scales, an approach that ensures that the

specification is provided in a modular way. Mappingsexist between NeuroML elements and several commonlyused simulators including NEURON [4], GENESIS [5] andPSICS [6], and a number of tools are available whichallow a user to create and validate NeuroML documentsand to generate code for model implementation by mul-tiple simulators from these documents. In particular, themodel development application neuroConstruct canimport and write NeuroML documents as well as generateoutput for simulating neuron or neuronal network activ-ity using either NEURON, GENESIS, PSICS or PyNN [7].Currently, NEURON can import and export cells in Neu-roML format, and import/export of NeuroML is in betatesting for PyNN, which is a Python package for simulatorindependent specification of neuronal network models[8]. The use of NeuroML with PyNN provides a connec-tion between the NeuroML descriptions of large-scaleneuronal network models and additional simulators.

Overall, NeuroML provides a valuable contributiontowards simulator interoperability as well as model pub-lication and exchange. The NeuroML standards will facil-itate a broad range of research goals in computationalneuroscience.

AcknowledgementsPartial support for this work was provided by the Wellcome Trust. We also thank the International Neuroinformatics Coordinating Facility and the

from Eighteenth Annual Computational Neuroscience Meeting: CNS*2009Berlin, Germany. 18–23 July 2009

Published: 13 July 2009

BMC Neuroscience 2009, 10(Suppl 1):L1 doi:10.1186/1471-2202-10-S1-L1

<supplement> <title> <p>Eighteenth Annual Computational Neuroscience Meeting: CNS*2009</p> </title> <editor>Don H Johnson</editor> <note>Meeting abstracts – A single PDF containing all abstracts in this Supplement is available <a href="http://www.biomedcentral.com/content/files/pdf/1471-2202-10-S1-full.pdf">here</a>.</note> <url>http://www.biomedcentral.com/content/pdf/1471-2202-10-S1-info.pdf</url> </supplement>

This abstract is available from: http://www.biomedcentral.com/1471-2202/10/S1/L1

© 2009 Crook et al; licensee BioMed Central Ltd.

Page 1 of 2(page number not for citation purposes)

BMC Neuroscience 2009, 10(Suppl 1):L1 http://www.biomedcentral.com/1471-2202/10/S1/L1

Publish with BioMed Central and every scientist can read your work free of charge

"BioMed Central will be the most significant development for disseminating the results of biomedical research in our lifetime."

Sir Paul Nurse, Cancer Research UK

Your research papers will be:

available free of charge to the entire biomedical community

peer reviewed and published immediately upon acceptance

cited in PubMed and archived on PubMed Central

yours — you keep the copyright

Submit your manuscript here:http://www.biomedcentral.com/info/publishing_adv.asp

BioMedcentral

National Science Foundation for support of NeuroML through workshop and travel funding.

References1. NeuroML Website [http://www.neuroml.org]2. Goddard N, Hucka M, Howell F, Cornelis H, Shankar K, Beeman D:

Towards NeuroML: Model description methods for collabo-rative modeling in neuroscience. Philos Trans R Soc Lond B Biol Sci2001, 356:1209-1228.

3. Crook S, Gleeson P, Howell F, Svitak J, Silver RA: MorphML: Level1 of the NeuroML standards for neuronal morphology dataand model specification. Neuroinformatics 2007, 5:96-104.

4. Hines ML, Carnevale NT: The NEURON simulation environ-ment. Neural Comp 1997, 9:1179-1209.

5. Bower JM, Beeman D: The Book of Genesis 2nd edition. New York:Springer-Verlag; 1998.

6. Parallel Stochastic Ion Channel Simulator (PSICS) Website[http://www.psics.org]

7. Gleeson P, Steuber V, Silver RA: neuroConstruct: A tool formodeling networks of neurons in 3D space. Neuron 2007,54:219-235.

8. Davison AP, Bruderle D, Eppler J, Kremkow J, Muller E, Pecevski D,Perrinet L, Yger P: PyNN: a common interface for neuronalnetwork simulators. Front Neuroinform 2009, 2:. doi: 10.3389/neuro.11.011.2008.

Page 2 of 2(page number not for citation purposes)