Understanding the Implications of Implementing FAIR data principles
21 Slides3.71 MB
Understanding the Implications of Implementing FAIR data principles in the Life Sciences Ebtisam Alharbi PhD researcher at the University of Manchester, UK [email protected] Prof. Carole Goble & Dr. Caroline Jay
Introduction Evidence Examine the implications of adopting FAIR in life sciences data. Establish a bridge between FAIR data principles and their implementations. Pass evidence to policy makers. FAIR data Life Sciences 2
FAIR Concepts A fundamental principles to make research data : Not having FAIR costs the European economy 10.2 billion (EC,2018) Wilkinson, M.D., et al., The FAIR Guiding Principles for scientific data management and stewardship. 2016. COMMISSION, E. 2018a. Cost-Benefit analysis for FAIR research data - Cost of not having FAIR research data.
The difficulty in understanding FAIR FAIR is FAIR is not A journey Machine readability of data and metadata Ambiguous Domain respectful A subset of indicators A standard Just about humans A specific Technology About open For specific domain Data quality or impact One size fits all Mons, B., et al., Cloudy, increasingly FAIR; revisiting the FAIR Data guiding principles for the European Open Science Cloud. 2017. 37(1): p. 49-56. Boeckhout, M., G.A. Zielhuis, and A.L. Bredenoord, The FAIR guiding principles for data stewardship: fair enough? European journal of human genetics, 2018. 26(7): p. 931. https://www.slideshare.net/carolegoble
FAIR Implementation Roadmap Define Concepts for FAIR implementation Economic Implement FAIR culture FAIR ecosystem Technical European Commission. (2018). Turning FAIR into reality Embed & Sustain Skills for FAIR Incentives and metrics for FAIR data and services Social Investment in FAIR Political
FAIR Assessment Metrics / indicators Automated VS Manual Degrees of FAIR https://github.com/FAIRMetrics/Metrics https://indico.cern.ch/event/588219/contributions/2384979/attachments/1426152/2188462/Dillo Doorn Assessing FAIRness CERN Geneva 13-0 3-2017-3.pd f
FAIRification process https://www.go-fair.org/fair-principles/fairification-process/
Implementing FAIR in Life sciences data https://elixir-europe.org https://fairplus-project.eu https://www.pistoiaalliance.org
A case study: The e-Tox Sampler Dataset e-TOX Partners A public access subset of the toxicological information manually compiled from the contents of pre-clinical studies as experimental result 13 pharma companies, 11 academia institutions, 6 SMEs The project life 2010 – 2016 Size http://www.etoxproject.eu 66 pre-clinical toxicology studies using 43 compounds
The challenges to FAIRify e-TOX Integration Extract and interlink data Manual intervention to determine data structure Ontology Selection of relevant ontologies Map to existing standards ontologies e.g. toxological Identifiers Determine alternative identifier strategies (InChi) accompanied by 2ndry IDs (ChEBI and CAS) Metadata Determine relevant community standards Vocabularies, and/or Schema.org or Bioschema
Pressing issues 1. How much does it cost? 2. What are the benefits? 3. Is it worth? 4. To what level of FAIRness? FAIR data
The Economic Implications of Implementing FAIR data principles in the Life Sciences Implication for Data Providers Implication for Funders
Economic Evaluation Landscape Economic Evaluation Cost Benefit Analysis (CBA) Cost Effectiveness Analysis (CEA) Cost Utilization Analysis (CUA) Cost Minimisation Analysis (CMA) Gertler, P.J., et al., Impact evaluation in practice. 2016: The World Bank.14. Robinson, R., Cost-benefit analysis. Bmj, 1993. 307(6909): p. 924-926. Boardman, A.E., et al., Cost-benefit analysis: concepts and practice. 2017: Cambridge University Press.
Early evidence of FAIR Cost SCHULTZ, O. R. A. H. B. 2018. Preliminary analysis: Introduction of FAIR data in Denmark. COMMISSION, E. 2018a. Cost-Benefit analysis for FAIR research data - Cost of not having FAIR research data.
What is Cost Benefit Analysis? Uncertainty about the actual impact and consequences. CBA is one of the standard evaluation technique (EC,2018). Applied in the social and economic impacts Pass evidence to policy makers. COMMISSION, E. 2018a. Cost-Benefit analysis for FAIR research data - Cost of not having FAIR research data.
Pre-FAIRification: Costs and Benefits Costs Time spent Data stewardship cost PIDs cost Licenses cost Storage cost Software cost Additional cost (e.g. long-term funding cost) Benefits Cost Savings Less time spent on data work for researchers New research produced Other likely benefits
Examples of Data stewardship Cost Number of data stewards Percentage of time dedicated to FAIRify this dataset Average salary (gross, yearly) Average working hours per week Average working days per year
CBA model: The roadmap for Pre-FAIRification
Conclusion Implementing FAIR in the Life sciences is challenging. Pass evidence to policy makers is a key. Cost Benefits analysis is the roadmap for the FAIRification process.
Acknowledgment
