At the Bio-IT World conference, held May 15-17 in Boston, the track on FAIR data saw a fair number of Dutch speakers present on the FAIR data initiative. Prof. Barend Mons, scientific director of the GO FAIR international support and coordination office, explained how he envisions the implementation of FAIR in daily practice.
Just to recapitulate for anyone who is not familiar with the initiative: FAIR aims to make scientific research data Findable, Accessible, Interoperable, and Reusable. And no, that’s not the way scientists currently handle their results. Most of them perform experiments and generate data with a publication in mind. When that article is published, the collected data is basically binned.
However, more and more scientists realise this approach is a waste of time and research funds. ‘It’s no exception that identical information is collected by different scientists to answer a similar research question.’
Mons would much rather see that scientists share their results: Reuse. This not only prevents redundancy in experiments; when researchers draw conclusions based on a larger number of observations, this also increases the credibility of the outcome.
Access instead of export
Mons stressed that sharing research data does not mean that the information needs to leave the company or the hospital. ‘Physicians and scientists are often not allowed to share patient information or confidential company data with third parties. Besides, the data packages could easily be too big to send to a personal computer.’
Mons, and others involved in the European GO-FAIR initiative that was launched earlier this year, therefore propose the development of the Internet of FAIR data and Services. This would imply that, just as with the regular internet, the data does not leave the pharmaceutical company, research institution or hospital. The scientist instead would be granted access to the data relevant to his or her research. Mons compared this situation with the railroad system. In his vision, the data is not transported from station A to station B, but the scientist (based at station B) is granted permission to stop and access the data at station A. To specify the terms of use, the parties involved could draw up a contract.
Government
A number of European governments, The Netherlands included, are actively promoting the generation of FAIR data and the reuse of results within the scientific community. ‘They no longer fund research proposals lacking a good stewardship plan, thereby promoting the creation of FAIR data.’
The FAIR data principles have been formulated in the past years, but how to implement those in daily practice is a question that has still not been fully answered. ‘We’re basically building the plane while flying it.’
Social change
Those technical issues need to be solved, and soon, Mons acknowledges. Yet he points out that IT is not the biggest hurdle. ‘FAIR requires a major social change from scientists: a change of focus from publications and citations to collaboration.’
Therefore, the first goal of the GO FAIR initiative is to facilitate the implementation of FAIR and to stimulate open science activities.
If you would like to know more on FAIR, contact us here.
