Big data from small data: data-sharing in the 'long tail' of neuroscience
Adam R Ferguson1, Jessica L Nielson1, Melissa H Cragin2
1Brain and Spinal Injury Center, Department of Neurological Surgery, University of California at San Francisco, San Francisco, California, USA.
Nature Neuroscience
|October 29, 2014
Summary
Sharing small neuroscience datasets, or long-tail data, is crucial for advancing research. Aggregating these diverse data sources creates big data, improving our understanding of brain disorders.
Area of Science:
- Neuroscience
- Neuroinformatics
- Scientific Data Sharing
Background:
- Growing international initiatives like the US BRAIN and European Human Brain Projects emphasize research transparency and data access.
- The increasing volume of data in neuroscience and a collaborative scientific environment necessitate robust data-sharing standards and infrastructure.
- Beyond 'big science,' individual neuroscientists face challenges managing and sharing diverse, heterogeneous small datasets.
Purpose of the Study:
- To address the critical issue of sharing small, heterogeneous neuroscience datasets, termed 'long-tail data.'
- To explore the utility, available repositories, and best practices for sharing these diverse data types.
- To demonstrate how aggregating long-tail data can generate valuable big data insights for neuroscience.
Main Methods:
- Commentary and analysis of current trends in neuroscience data sharing.
- Review of existing data repositories and sharing options for individual researchers.
- Presentation of use cases illustrating the aggregation of small datasets into big data.
Main Results:
- Long-tail data, though small individually, holds significant potential when aggregated.
- Diverse repositories and evolving best practices support the sharing of heterogeneous neuroscience data.
- Aggregating long-tail data transforms numerous small sources into a powerful big data resource.
Conclusions:
- Effective sharing of long-tail neuroscience data is essential for collaborative research and advancing knowledge.
- The strategic aggregation and mining of small datasets can yield significant insights into neuroscience-related disorders.
- Developing neuroinformatics infrastructure and data-sharing standards is paramount for the future of neuroscience research.
