The value of data sharing

Added by Branwen Hide on 02 March 2009 18:01

Research data are fast being recognised as valuable resources in their own right, as opposed to being secondary to scholarly outputs such as a journal articles.

One of the pushes behind this movement is the array of technological developments that have allowed researchers to generate increasing amounts of digital data (see Big Data). Another is the growing recognition of the potential value of making research data available for sharing by researchers, research funders and the UK government. The UK government?s Science and Innovation Investment Framework 2004-2014, argues that the UK research base should have ready and efficient access to digital information of all kinds. As valuable resources, data need to be maintained and, where appropriate, restructured (addition of metadata) to allow easy retrieval, long term preservation and re-use. Though technologies which support these new opportunities continue to evolve at a rapid pace, researchers? attitudes to data creation and dissemination are not necessarily keeping pace in all disciplines.

Over the past few weeks I have been to a couple of conferences looking at various aspects of data sharing. The first was the NERC Data management workshop, and from the point of view of the RIN, it was quite nice to be in a room full of the converted. NERC have a number of longstanding data centres, and a data policy to support them. One of the objects of the meeting was to determine the future of NERC data management with the aim to fit this into the NERC Science Information Strategy. However, despite the widespread support for data sharing in the room, many of the attendees talked about the difficulty in getting some researches to actively engage and share their data in a meaningful way. The Centre of Ecology & Hydrology has specialist staff members who liaise between those looking after the data centres and the researchers.

Furthermore, the UK Research Data Service (UKRDS) had a conference last week, to talk about the findings of the study, which include the development of a pathfinder project to examine the feasibility of developing a national shared service for managing research data. There was also a lot of discussion as to how one determines the value of data sharing in both economic and non-economic terms. Sharing data has a potential to open new areas of research/questioning that the sharer had not thought of before. Though there are obvious examples of where large data centres have enormous value (such as the EBI, the UKDA and the NERC data centres), what will often convince researchers more are the small-scale projects which have adapted the large-scale technologies.

Recently researchers led by Dr Grant Churchill used computational methods to create a cheap and efficient drug discovery system that will allow small labs to search large databases just as pharmaceutical companies do. ROCs was provided for free and enabled them to search the freely available ZINC data base for compounds which correspond to small molecules instead of the traditional and time-consuming method of starting with a chemical of interest and systematically modifying its chemical structure. This is just one of many small examples of the benefits of data sharing, but I am sure there are others out there, so if you know of any I would be happy to hear about them.  

© Research Information Network 2005–2009