New publicly funded research methods in the life sciences, such as high-throughput screening for drug discovery, are producing increasingly large volumes of data. That in turn sharpens the need to make these data as widely available as possible in the public interest.
These considerations underpin a policy statement on data-sharing published by the Biotechnology and Biological Sciences Research Council (BBSRC), the government-sponsored agency that invests around £370 million per year in life sciences research in the UK.
Following a web-based consultation with the research community in 2005/06, the BBSRC has set out its expectations on data-sharing and best practice in the field. Under the policy, all scientists applying for research grants from the Council will need to provide a detailed statement on research data-sharing, which will be assessed along with the rest of their application.
“Our new policy makes it clear that we expect scientists to share data as an integral part of their research,” commented BBSRC chief executive Julia Goodfellow. “And we will provide funds to enable them to do this.”
According to the policy statement, the BBSRC wants research data generated as a result of its support to be “made available with as few restrictions as possible in a timely and responsible manner to the scientific community for subsequent research”. Applicants for BBSRC grants should make data available through existing community resources or databases where possible, it adds. Data should also be retained for 10 years after completion of a research project in line with the BBSCR Statement on Safeguarding Good Scientific Practice.
Researchers have a legitimate interest in benefiting from their own time and effort in producing data, the policy acknowledges. That interest does not extend, though, to prolonged exclusive use of those data.
“Timescales for data-sharing will be influenced by the nature of the data, but it is expected that timely release would generally be no later than the release through publication of the main findings and should be in line with established best practice in the field,” the document says. Where no best practices exist, data-sharing within three years of the dataset being generated is suggested as a guide.
The BBSRC also supports the view that researchers who enable data-sharing should “receive full and appropriate recognition by funders, their academic institutions and new users for promoting secondary research”.
Data quality and provenance are another important consideration, the policy notes. Where appropriate, it says, shared data should be accompanied by contextual information or documentation providing secondary users with any necessary details on the origin and manipulation of the data, so as to prevent any misuse, misinterpretation or confusion.
The BBSRC says it recognises that effective data-sharing already occurs in certain areas and it “expects this to continue”. However, the consultation process identified two areas where there was a “particularly strong scientific case” for sharing. These involved data arising from high-volume experimentation; and low-throughput data arising from long-time series or cumulative research approaches.
The new policy aims to achieve data-sharing “in an appropriate manner”, without being too prescriptive. All the same, the BBSRC reserves the right to implement a more prescriptive approach, particularly in cases involving large-scale community research or where the Council is supporting a community resource.