January 14, 2016 - 10:00am PST
Electrophysiology Data Discovery Index (EDDI) - Raimond L. Winslow, PhD
Dynamic data are produced by sampling continuous-time waveforms to generate time-series. They are ubiquitous throughout biomedical research. Examples include the electrocardiogram, other patient monitoring data, and single- and multi-electrode whole-cell and patch-clamp recording data. Unlike the case for static data, there is no culture of data sharing, standards and general purpose tools for publishing and discovering dynamic biomedical data. Currently, these data are either lost forever or stored on digital media in laboratories of individual investigators where they at best must be accessed manually, or at worst can’t be accessed at all. In light of the huge investment made by NIH to collect these data, it is remarkable that such a broad class of biomedical data is “ephemeral”.
To help address this problem, in a supplement to the CardioVascular Research Grid (CVRG), we have developed the Electrophysiology Data Discovery Index (EDDI), a web-based platform for annotating, publishing, and discovering biomedical time-series data. Built using DSpace, EDDI supports the annotation, publication and discovery of electrophysiology time-series data. Users publish data by entering the URLs where data sets are located in web-forms, annotate data sets, and associate data sets with peer-reviewed publications. EDDI uses Apache Solr to index these entries and provides a search interface to aid data discovery. EDDI has integrated Globus OAuth authentication which allows researchers with a Globus Online account to quickly register to use EDDI using their existing Globus credentials. This Globus integration enables EDDI users to quickly reference data repositories defined as Globus endpoints, and seamlessly integrate with Globus Connect to initiate high-speed data transfers. While tailored to the needs of the electrophysiology data communities, EDDI is designed to allow for the addition of new dataset communities and also to individual community dataset schemas as the needs of researchers evolve. We will do a live demonstration of the functionality provided by EDDI.
Raimond L. Winslow, Ph.D., The Johns Hopkins University School of Medicine & Whiting School of Engineering
Dr. Winslow is the Raj and Neera Singh Professor of Biomedical Engineering, and Director of the Institute for Computational Medicine at Johns Hopkins University. Work in his laboratory is focussed in two areas. The first is understanding the relationship between the perturbed function of molecular networks and generation of arrhythmia in heart disease through development and application of computational models. The second is developing informatics tools that enable sharing and analysis of cardiovascular clinical research data. He is PI of the CardioVascular Research Grid Project (cvrgrid.org), an NHLBI-funded R24 resource developing web-based tools in this area.