gms | German Medical Science

Introduction: Recent fundamental enhancements in advanced light microscopy (ALM) raises new opportunities to answer biomedical research questions [1]. However, the speed of technological advances clearly outperforms an adequate management of the resulting research data. Thus, there is an urgent need for sustainable (IT) infrastructures to improve the reusability of images and pertinent metadata after data publication [2].

Integrating key concepts of Linked Open Data [3], FAIR guiding principles [4], and Research Data Alliance (RDA) [5] in the collaborative research center (CRC) 1002 and in German Centre for Cardiovascular Research (DZHK) we aim at providing a sustainable research data management module for microscopy data as part of our Research Data Platform (RDP) [6]. Here we describe a web-based implementation of requirements from researcher and IT perspective as well as recommendations for the documentation of essential metadata in order to develop a sustainable ALM research data management.

Methods: To determine typical and essential metadata in ALM we analyzed selected experimental workflows for two ALM microscopes used in CRC 1002 and DZHK-associated experiments. In close cooperation with CRC 1002 and DZHK microscopy experts we determined requirements to enhance reproducibility and subsequent reuse of microscopy derived datasets. Based on this analysis, we developed a metadata schema which was implemented as an ALM RDP web service module for facilitated metadata acquisition and management.

Results: Our requirements and workflow analysis revealed that ALM metadata can be divided into two main categories: the acquisition metadata and the experimental metadata. The acquisition metadata contains information about the microscope settings during the acquisition e.g. stimulation wavelength, used lenses or the type of image complexity. The experimental metadata include mainly descriptions about the experimenter, the scientific question, the samples and their provenance, and the staining method (e.g. antibodies and fluorescence dyes). While the majority of acquisition metadata is automatically generated by the microscope instrument software, the experimental metadata have to be added by the researchers. Initially, the microscope software creates image data in proprietary file formats which contain the acquisition metadata. Our analysis demonstrates that data conversion into the open ome-tiff image format does not optimally transport all metadata values. Based on our developed metadata schema, our ALM RDP module was implemented as a web service and is currently in use for CRC 1002 metadata documentation.

Discussion: ALM technologies are rapidly evolving in terms of technical diversity and massively increasing data volumes. We present a web-based approach to facilitate metadata acquisition which is initially implemented for scientists in the context of CRC 1002 and DZHK. Current alternatively available commercial imaging databases are usually hardly affordable in academic settings and lack compatibility with the FAIR principles. Open source tools like OMERO [7] rather focus on the documentation of individual images and only partially reflect their embedding into complex scientific workflows. In summary, the high demand for research data management solutions in ALM-based science [2] bears a great potential for the reuse of our ALM module in other research settings.

Acknowledgements:

This work was funded by the German Research Foundation (DFG) within the grant for the Collaborative Research Centre (SFB) 1002 on Modulatory Units in Heart Failure, subproject INF and the German Federal Ministry of Education and Research (BMBF) within the framework of the German Center for Cardiovascular Research (DZHK) (FKZ 81Z7300173).

Die Autoren geben an, dass kein Interessenkonflikt besteht.

Die Autoren geben an, dass kein Ethikvotum erforderlich ist.