Publishing research data without effort: benefits of cooperation between the UR Data Hub and scientists in the immediate and automatic publication of data - University of Regensburg Publication Server

Abstract:  The workflow for the publication of research data on the institutional repository of the University of Regensburg (UR) has been developed to enable the researchers to make data available in an immediate and automated manner. The development of this workflow has been undertaken by the UR Data Hub, with its current testing being carried out by the Department of Neurogenetics at the UR.

The UR Data Hub is the central institution for research data management (RDM) at the UR [1]. The UR Data Hub's primary objective is the establishment and further development of infrastructures on a university-wide scale, with the aim of facilitating the daily RDM activities of researchers. Requests for new or further developed services made by the scientists at the UR are discussed and reviewed by the RDM Board. The implementation process is conducted at a centralised level. Depending on the degree of effort required for the request, scientists may be granted membership of the UR Data Hub, thus enabling them to collaboratively develop the requested workflow with the RDM experts. The philosophy of the UR Data Hub is the provision of the implemented service to not solely the scientist, who made the request, but to all members of the university. This approach enables the consideration of subject-pecific requirements (e.g. by the project funding agencies). One request that has recently been made is the immediate and automated deposit of research data into the institutional repository. The aim is to make biological experimental data openly available at every stage of the data collection process, e.g. from the very beginning [2]. The research data contains a configuration file including metadata in a structured and machine readable format. The metadata has to be provided by the scientist at the begining of the experiment describing the whole setup of the experiment. A Python script executes the data upload. When the research data is pushed to the repository, the metadata from the configuration file is used to generate findable and interoperable data in the repository entry, including the chosen licence for re-using the data. It is intended to publish new research data on a regular basis (in this case daily), which will update the data already published. Therfore, the data is available at the time when it is measured. On demand a DOI is assigned to a record and therefore a citable dataset will be produced. This allows the data to be reused even if the experiment is not yet finished. The flexible structure of the configuration file enables the definition of various file types and metadata depending on the subject and eperimental setup. Consequently, the automated publication workflow of datasets can be utilised and further developed for scientists from other disciplines. This contribution will provide a comprehensive overview of the automated deposit process, encompassing the initial conceptualisation of the process and its subsequent implementation at the UR. It also introduces the UR Data Hub and its organisational structure.