Skip to main content

FAIRplus

Periodic Reporting for period 1 - FAIRplus (FAIRplus)

Reporting period: 2019-01-01 to 2020-06-30

Wide sharing of data drives the progression of science. The wealth of data and knowledge generated from a €5.3bn public-private investment and over 100 IMI projects, spanning all areas of pharmaceutical R&D, provides a huge opportunity to advance European science; it is a foundation of reference data that can be exploited by researchers for decades. Linking data from different domains brings advancements in research and discovery; for drug discovery, a greater understanding of the mechanisms of disease, infection and chemistry brings improvements in decision-making, reproducibility and drug development success rates. Yet the IMI data is often stored in project-specific proprietary databases, that in isolation provide little opportunity for reuse either outside of the project, or beyond the duration of the project. Inconsistent annotation between projects - and in internal EFPIA participant databases - blocks aggregation to large data reservoirs and is a major hurdle for integration, knowledge mining and advanced modelling.

FAIRplus (fairplus-project.eu/) is developing tools and guidelines for making life science data FAIR (Findable, Accessible, Interoperable, Reusable). Through worked examples using IMI and EFPIA data and application and extension of existing methods, we will improve the level of discovery, accessibility, interoperability and reusability of selected IMI and EFPIA data. In addition, through disseminated guidelines and tailored training for data handlers in academia, SMEs and pharmaceuticals, data management culture will change and be sustained, and datasets will be reused by pharmaceutical companies, academia and SMEs. Our FAIR Innovation and SME programme (https://bit.ly/3m44Vz4) will enable wide data reuse and foster an innovation ecosystem around these data that power future re-use, knowledge generation, and societal benefit.
During the first 18 months we have:

Identified project data sources for FAIRification:
A process (including a survey and ethics review) for selecting and prioritising IMI project databases for FAIRification has been established (https://bit.ly/3jZHMfp) which involved agreeing upon criteria, including societal impact, to prioritise and select IMI projects with data most suitable for FAIRification. This process has been applied to 25 IMI projects, with more than 10 completed surveys. Four pilot projects were selected, with three datasets made publicly available and one made available to the FAIRplus consortium only (https://bit.ly/2GGnFo9). Transcriptomic data has been prioritised as a data type of highest interest to EFPIA partners.

Developed initial tools and metrics needed to make data Findable, Accessible, Interoperable and Reusable (FAIR):
Two main tools (FAIR CMMI and FAIR cookbook) are in development that will enable researchers to understand how to assess the FAIR level of datasets, assess the benefits of achieving a higher level of FAIR and follow a process and guidelines on how to actually make data sets more FAIR. The FAIR cookbook (https://bit.ly/2DIEWMt) has recipes that include templates, new recipes and examples where the recipes are used to enable the FAIRification of datasets. There are eight recipes in the published version and 38 recipes in the development version that will soon be made public. Four of the published recipes within the FAIR cookbook are from the selected IMI datasets and one is published in Scientific Data. Tools have also been developed to ensure that an ethics procedure is implemented throughout FAIRplus. Use of these tools ensures measures are followed, which the various persons involved in data stewardship, data governance and data use, need to document and respect when an IMI project agrees to give access to their data.

Developed programmes, communications and events to change and sustain the data management culture across academia, SMEs and pharmaceuticals:
To change and sustain data management culture we have taken a multifaceted approach:
1) Broadly communicated FAIRplus to update the wider community (https://twitter.com/FAIRplus_eu https://www.linkedin.com/company/40786325/admin/);
2) Developed and run a FAIRplus Innovation and SME Forum (https://bit.ly/3m44Vz4) to engage with SMEs that have an interest in FAIR data processes who may be able to sustain activities beyond the end of FAIRplus;
3) Built an execution plan in order to run the FAIRplus Fellowship Programme (https://bit.ly/35lC4R0) which will launch in 2021. The Fellowship Programme aims to educate the next generation of experts for further FAIRification of data sets within IMI projects, EFPIA partners and beyond. During the programme, fellows will improve the FAIR levels of their own data sets, adopt the cookbook recipes together and learn how to apply the CMMI model. To date, course curriculum and advertising materials have been the focus;
4) Developed plans to engage with policy makers;
5) Considered options to sustain outputs of the FAIRplus project beyond the project itself.
Data management plans are now a standard component of project proposals but yet there is still little understanding of the resources required or steps to take to maximise the value and impact of data beyond the specifics of that particular project. In addition, pharmaceutical companies and SMEs often have internal silos of data that are not managed effectively. The FAIRplus project will bring about a change in the way data scientists can approach the data and value of the data they produce. Scientists are either unaware of what the FAIR principles are or are not equipped with the knowledge to implement them.

By the end of FAIRplus, we will have driven a cultural change in data management that will be long lasting within academia and industry, including the following:
- We will have delivered operational and scalable processes that can be applied to assess the value and decision making to make the FAIR principle work. Tools and recipes will guide data scientists throughout the process.
- Data from 20 IMI projects will have been taken to a higher level of FAIR and therefore can provide impact by increased data reuse.
- We will have worked closely with SMEs to look for opportunities to grow the SME market e.g. combining commercial data that cannot be shared with public data.
- We will have contributed to a cohesive tools landscape by using tools that are already available, by reusing code, improving the interoperability between tools and reducing costs of redevelopment and tool installation.
- We will have expanded the FAIR community in Europe, through dissemination of FAIRplus activities, promotion of FAIRification as a research area, and training of researchers to improve capacity to promote FAIR data generation.

The FAIRplus project overall has increased the uptake and implementation of FAIR principles and the future sustainability of data. By prioritising datasets with the potential to generate high socio-economic impact (promoting healthy aging, addressing chronic diseases, neurodegenerative diseases and the emergence of antibiotic resistance), we have ensured the pilots used to develop the FAIR cookbook and actual data FAIRification has been of greatest value. This selection process will continue to ensure scientific data from selected IMI projects and EFPIA are broadly usable. In addition, the FAIR-CMMI will demonstrate the value of this approach and required culture change to enable further uptake.
FAIRplus