Considerations for developing a standard for storing electrophysiology data in HDF5
Jeffrey
L.
Teeters1*,
Jan
Benda2,
Andrew
P.
Davison3,
Stephen
Eglen4,
Stephan
Gerhard5,
Richard
C.
Gerkin6,
Jan
Grewe2,
Kenneth
Harris7,
Tom
Jackson8,
Roman
Mouček9,
Robert
Pröpper10,
Hyrum
L.
Sessions11,
Leslie
S.
Smith12,
Andrey
Sobolev2,
Friedrich
T.
Sommer1,
Adrian
Stoewer2 and
Thomas
Wachtler2
-
1
University of California at Berkeley, United States
-
2
Ludwig Maximilian University, Germany
-
3
Centre Nationale de la Recherche Scientifique, France
-
4
University of Cambridge, United Kingdom
-
5
ETH Zurich, Switzerland
-
6
Carnegie Mellon University, United States
-
7
Imperial College London, United Kingdom
-
8
University of York, United Kingdom
-
9
University of West Bohemia, Czechia
-
10
Technical University of Berlin, Germany
-
11
Blackrock Microsystems, United States
-
12
University of Stirling, United Kingdom
The INCF Program on Standards for Data Sharing has a working group that is developing a standard for storing electrophysiology data in HDF5. The impetus for this effort is that many experimentalists are starting to use HDF5 to store data, so a standard would facilitate data sharing significantly.
The most important requirement of such a standard is to accommodate the common types of data used in electrophysiology and also the metadata required to describe them. Neuroshare (an API for accessing electrophysiology data stored in various formats) defines four data types: analog signals, segments, neural events and experimental events; as well as some metadata. A standard needs to efficiently store these data types, and probably also imaging data and some kinds of data generated in the data processing chain, such as features used for spike sorting.
Further, a standard way of storing the metadata must be specified. The set of metadata required to describe electrophysiology data is difficult to determine a priori because the types of experiments are so varied. So, a flexible mechanism must be used which allows referencing and specifying values for currently existing ontologies and also accommodates information not currently systematized. Techniques to include post-experiment annotations of data, and for relating different data parts, are also required.
There are numerous projects relevant to storing electrophysiology data in HDF5. These include: NEO, NeuroHDF, brainliner.jp, klusta-team spikedetekt, BrainVisionHDF5 and Ovation (ovation.io). A project, NeXus Format (nexusformat.org), uses HDF5 to store particle physics data, but might be adaptable for electrophysiology. It is managed using a well-defined community infrastructure that may be worth emulating.
So far, the working group entertains two approaches towards defining a standard, which may eventually be merged. One, currently named Pandora, defines a generic data model that can be used with HDF5 or other storage back-ends. Due to the generic nature, the data model can be used to store various kinds of neuroscience data. The other proposal, called epHDF, defines domain specific schemata for storing electrophysiology data in HDF5. For any approach, a suite of test data sets to help evaluate a proposed standard is needed, and tools to allow validating data files are desirable.
Details of the above considerations and the current state of the development of a standard will be presented.
Acknowledgements
Acknowledgment: This work was conducted within the Electrophysiology Task Force of the INCF Program on Standards for Data Sharing. Funding for J.L. Teeters and F.T. Sommer provided through NSF grant 0855272.
Keywords:
HDF5,
data sharing,
metadata,
Electrophysiology,
Standards
Conference:
Neuroinformatics 2013, Stockholm, Sweden, 27 Aug - 29 Aug, 2013.
Presentation Type:
Poster
Topic:
Electrophysiology
Citation:
Teeters
JL,
Benda
J,
Davison
AP,
Eglen
S,
Gerhard
S,
Gerkin
RC,
Grewe
J,
Harris
K,
Jackson
T,
Mouček
R,
Pröpper
R,
Sessions
HL,
Smith
LS,
Sobolev
A,
Sommer
FT,
Stoewer
A and
Wachtler
T
(2013). Considerations for developing a standard for storing electrophysiology data in HDF5.
Front. Neuroinform.
Conference Abstract:
Neuroinformatics 2013.
doi: 10.3389/conf.fninf.2013.09.00069
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:
09 Apr 2013;
Published Online:
11 Jul 2013.
*
Correspondence:
Dr. Jeffrey L Teeters, University of California at Berkeley, Berkeley, United States, teeters@berkeley.edu