gms | German Medical Science

66. Jahrestagung der Deutschen Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie e. V. (GMDS), 12. Jahreskongress der Technologie- und Methodenplattform für die vernetzte medizinische Forschung e. V. (TMF)

26. - 30.09.2021, online

The bumpy road of FAIRification in practice

Meeting Abstract

  • Harald Kusch - Institut für Medizinische Informatik, Universitätsmedizin Göttingen, Göttingen, Germany; Cluster of Excellence “Multiscale Bioimaging: from Molecular Machines to Networks of Excitable Cells” (MBExC), University of Göttingen, Göttingen, Germany; Campus-Institute Data Science (CIDAS), Göttingen, Germany
  • Christian R. Bauer - Institut für Medizinische Informatik, Universitätsmedizin Göttingen, Göttingen, Germany
  • Theresa Bender - Institut für Medizinische Informatik, Universitätsmedizin Göttingen, Göttingen, Germany
  • Jonas Hügel - Institut für Medizinische Informatik, Universitätsmedizin Göttingen, Göttingen, Germany; Campus-Institute Data Science (CIDAS), Göttingen, Germany
  • Robert Kossen - Institut für Medizinische Informatik, Universitätsmedizin Göttingen, Göttingen, Germany
  • Sara Yasemin Nussbeck - Institut für Medizinische Informatik, Universitätsmedizin Göttingen, Göttingen, Germany; Universitätsmedizin Göttingen, Zentrale Biobank, Göttingen, Germany
  • Ulrich Sax - Institut für Medizinische Informatik, Universitätsmedizin Göttingen, Göttingen, Germany; Campus-Institute Data Science (CIDAS), Göttingen, Germany

Deutsche Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie. 66. Jahrestagung der Deutschen Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie e. V. (GMDS), 12. Jahreskongress der Technologie- und Methodenplattform für die vernetzte medizinische Forschung e.V. (TMF). sine loco [digital], 26.-30.09.2021. Düsseldorf: German Medical Science GMS Publishing House; 2021. DocAbstr. 169

doi: 10.3205/21gmds072, urn:nbn:de:0183-21gmds0728

Veröffentlicht: 24. September 2021

© 2021 Kusch et al.
Dieser Artikel ist ein Open-Access-Artikel und steht unter den Lizenzbedingungen der Creative Commons Attribution 4.0 License (Namensnennung). Lizenz-Angaben siehe http://creativecommons.org/licenses/by/4.0/.


Gliederung

Text

Introduction: In general, requirements for collaborative research data management (RDM) should be similar in basic research, translational research, and in clinical research. However, in reality we face very different details regarding the needs of biomedical research consortia like Collaborative Research Centers (CRC), or Clinical Research Units (CRU). These differences can be assessed along the axes of (I.) Data types and data policy, (II.) Workflows, (III.) Privacy and (IV.) Tools. Implementing FAIR data management aspects (“FAIRification”) [1], e.g. in the context of patient derived xenografts in a CRU, requires the determination of relevant (meta-)data formats as well as agreements on their sharing policies. Usually this needs to be approached by a portfolio of different tools across several network security segments. Here, we describe our experience on the most “bumpy” aspects that make the FAIRified road construction challenging.

Methods: In different project contexts, ranging between basic, translational, and clinical research (e.g. CRU 5002, CRC 1002, Molecular Tumor Board Report), we are applying similar processes along the four described axes and the FAIR guiding principles to facilitate, organize, and standardize research data acquisition, integration and visualization [2], [3], [4], [5]. This involves aspects like data stewardship, tool development and adaption, data review and data quality assessment.

Results: Various tasks along the four described axes, especially selection of relevant (meta-)data types, were greatly facilitated by the close collaboration with selected early adopters and service facilities. However, a variety of organizational and technical issues usually come up during implementation of data management processes and tools across the different use cases: (I.) Selection of data types: Experimental approaches and appropriate sharing modes in the research consortia were highly dynamic and heterogeneous. The selection processes and data policies require prioritization of the most important FAIRification options. The data types combined with the data policy and the workflows determine whether the data was stored in a central repository or just cross-linked to this repository. (II.) Workflow representation: According to the variety of data types, also workflows were often complex and dynamic. Workflow management processes are often not clearly defined and consented and need to be formalized before tools can be applied for workflow documentation and visualization. (III.) Privacy: The privacy demands of the data types partially require special security zones for the modules of the data management framework. (IV.) Tools: A “Swiss army knife” of FAIRification tools was not available on the market or in the scientific communities. Portfolios of different tools have to be reused and adapted or developed and integrated in the different consortial infrastructures.

Discussion: Although on a higher abstraction level FAIRification requirements are comparable in diverse biomedical research consortia, details in (meta-)data, data policies, privacy concerns and suitable tools are heterogeneous and highly dynamic. As a consequence, a “Swiss army knife” of FAIRification is neither available for RDM requirements engineering nor for corresponding tools. In addition, a lack of standardization of processes, tools and interfaces as well as a lack of professional staff education (such as data stewards or software developers) still make FAIR RDM challenging.

Acknowledgement: This project was funded partially by: the DFG through projects KFO5002 (project 426671079), myPathSem (i:DSem 031L0024A), MTB Report (VW Stiftung), CRC 1002 infrastructure (INF) project, as well as the Z projects of CRC 1190 and CRC 1286, and Germany’s Excellence Strategy - EXC 2067/1- 390729940.

The authors declare that they have no competing interests.

The authors declare that an ethics committee vote is not required.


References

1.
GO FAIR. FAIRification Process [Internet]. 2019. Available from: https://www.go-fair.org/fair-principles/fairification-process/ Externer Link
2.
Bauer CR, Umbach N, Baum B, Buckow K, Franke T, Grütz R, et al. Architecture of a Biomedical Informatics Research Data Management Pipeline. Stud Health Technol Inform. 2016;228:262–6.
3.
Bender T, Bauer CR, Parciak M, Lodahl R, Sax U. FAIR conform ETL processing in translational research. In: Deutsche Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie, editor. 63 Jahrestagung der Deutschen Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie e.V. (GMDS). Osnabrück, 02.-06.09.2018. Düsseldorf: German Medical Science GMS Publishing House; 2018. DocAbstr. 254. DOI: 10.3205/18gmds095 Externer Link
4.
Bauer CR, Knopp C, Bender T, Kusch H, Sax U. Application of basic research data management with FAIRDOM/SEEK from a medical informatics perspective. In: Deutsche Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie, editor. 63. Jahrestagung der Deutschen Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie e.V. (GMDS). Osnabrück, 02.-06.09.2018. Düsseldorf: German Medical Science GMS Publishing House; 2018. DocAbstr. 205. DOI: 10.3205/18gmds093 Externer Link
5.
Suhr M, Lehmann C, Bauer CR, Bender T, Knopp C, Freckmann L, et al. Menoci: lightweight extensible web portal enhancing data management for biomedical research projects. BMC Bioinformatics. 2020;21(1):582.