This is a presentation of the discussion session "Streamlining Data Publication: Automatic Metadata and Large Datasets in the Age of AI" created by the DiTraRe team for the Open Science Conference in Hamburg 2025.
Research data repositories are essential infrastructure for enabling Open Science and ensuring data is Findable, Accessible, Interoperable, and Reusable (FAIR). However, researchers working on repositories face significant challenges in handling ever-increasing volumes of large datasets and the often time-consuming manual process of creating comprehensive, quality metadata. These issues can hinder data publication workflows and limit the findability and usability of valuable research output.
We will present the challenges encountered, propose solutions and first implementations for automatic metadata extraction and large data handling, and discuss how these innovations contribute to a more streamlined and scalable data publication workflow. Participants will have the opportunity to engage with developers and users of repositories, explore the practical implications for their own data management practices, and discuss the potential for adopting similar solutions in other repository contexts.
This discussion session will provide insights into the approaches, technologies, and lessons learned from the Leibniz Science Campus “Digital Transformation of Research” (DiTraRe) work on RADAR and implementation of AI methods in the topic of metadata standardisation. The session is especially relevant for everyone interested in the practical implementation of advanced research data management features that promote reproducibility, efficiency and FAIR principles.
Artificial intelligence, and specifically LLMs, have the capability to organize scholarly knowledge and support researchers throughout the research life cycle. In this presentation, we will discuss two AI services developed by TIB, that provide researchers with tools to leverage AI in their work.
Firstly, we will present ORKG Ask, a scholarly search and exploration platform. ORKG Ask allows users to ask research questions and find literature related to these questions among it more than 75 million indexed scholarly artifacts. Information is automatically extracted to answer the research question and is presented alongside additional key insights from the articles. A summary of the most relevant literature is generated as well. Finally, filtering tools allow users to narrow down the search to find specific articles they are interested in.
Secondly, we discuss the TIB AIssistant, a novel service that provides researchers AI support throughout the research life cycle. The AIssistant is a domain agnostic platform where different assistants can be used in sequence to perform specific research tasks. In addition to the specialized prompts, it integrates external scholarly services that are organized in the tool library. Examples of assistants include research idea generation, research question formulation, literature discovery, and article drafting. Ultimately, the AIssistant provides researchers with a flexible AI-supported platform for accessing tools and approaches to support their research.
Poster "Strengthening Research Data Management in Sport Science: A Community-Based Approach to Guidelines, Practices, and Challenge" at the Data Stewardship goes Germany 2025 Workshop at the KIT.
Präsentation zum Thema "Aufbau einer Support-Struktur für das Forschungsdatenmanagement: Das Research Data Center Motor Performance" im Arbeitskreis "Forschungsdatenmanagement in der Sportwissenschaft: Aktuelle Entwicklungen" im Rahmen des dvs Hochschultag 2025 in Münster.
This presentation given at Computing in Cardiology 2025 in São Paulo, Brazil on Sep 17 presents an overview of the opinion paper Computational modelling of biological systems now and then: revisiting tools and visions from the beginning of the century, which appeared in Philosophical Transactions A Volume 383, Issue 2296 (doi:10.1098/rsta.2023.0384).
The Leibniz ScienceCampus "Digital Transformation of Research" (DiTraRe), a collaborative initiative by FIZ Karlsruhe and Karlsruhe Institute of Technology (KIT), develops and pilots trusted digital research workflows to enhance reproducibility, transparency, and reusability in the natural sciences. Addressing the challenges posed by increasing data volumes and AI integration, DiTraRe focuses on four interdisciplinary research clusters:
- Protected Data Spaces: Develops methods for securely handling sensitive data in sports science, ensuring privacy protection while enabling research.
- Smart Data Acquisition: Optimizes digital data collection and analysis in chemistry through tools like the Chemotion Electronic Lab Notebook, aiming for efficient and standardized data capture.
- AI-Based Knowledge Realms: Investigates the application of AI in biomedical engineering, such as predicting intensive care unit stays, emphasizing explainability and reliability of AI methods.
- Publication Cultures: Explores new models for publishing large datasets, particularly in climate research, to improve accessibility and reuse of scientific data.
DiTraRe leverages the modular infrastructure of RADAR, FIZ Karlsruhe’s established repository platform, to support secure data publication, workflow versioning, and long-term preservation. Funded by the Leibniz Association, the project exemplifies how interdisciplinary collaboration and robust infrastructure can foster trust in digital research across various scientific domains.
Our poster presents DiTraRe’s architecture, pilot cases, and development roadmap, demonstrating practical approaches to institutionalizing trust in scientific workflows.
In response to the growing importance of large language models (LLMs) and big data workflows, we have developed a secure, locally hosted AI platform at the Leibniz Institute for Astrophysics Potsdam (AIP). The service, available under https://ai.aip.de, enables all AIP researchers and staff to access advanced LLM capabilities directly—without relying on external providers.
Built entirely on open-source technologies and hosted on AIP’s infrastructure, the platform prioritizes data security, low-latency interaction, and customizability. Users can leverage the system for a wide range of tasks including text generation, code assistance (“vibe coding”), and scientific data workflow generation.
A standout feature is our integration with the REANA reproducible research platform. An in-house agentic LLM assistant can autonomously analyze a scientific query, retrieve datasets, write code, and build a complete REANA workflow pipeline. This pipeline—comprising source code, configuration, plots, and metadata—is executable directly on AIP servers.
This innovation opens new possibilities for AI-assisted science. By automating data analysis tasks and integrating tightly with AIP’s computing ecosystem, ai.aip.de helps researchers focus on scientific discovery while AI handles routine and technical complexity.
Chemistry is an example of a discipline where the advancements of technology have led to multi-level and often tangled and tricky processes ongoing in the lab. The repeatedly complex workflows are combined with information from chemical structures, which are essential to understand the scientific process. An important tool for many chemists is Chemotion, which consists of an electronic lab notebook and a repository. This paper introduces a semantic pipeline for constructing the BFO-compliant Chemotion Knowledge Graph, providing an integrated, ontology-driven representation of chemical research data. The Chemotion-KG has been developed to adhere to the FAIR (Findable, Accessible, Interoperable, Reusable) principles and to support AI-driven discovery and reasoning in chemistry. Experimental metadata were harvested from the Chemotion API in JSON-LD format, converted into RDF, and subsequently transformed into a Basic Formal Ontology-aligned graph through SPARQL CONSTRUCT queries. The source code and datasets are publicly available via GitHub. The Chemotion Knowledge Graph is hosted by FIZ Karlsruhe Information Service Engineering. Outcomes presented in this work were achieved within the Leibniz Science Campus ``Digital Transformation of Research'' (DiTraRe) and are part of an ongoing interdisciplinary collaboration.
Despite the growing availability of tools designed to support scholarly knowledge extraction and organization, many researchers still rely on manual methods, sometimes due to unfamiliarity with existing technologies or limited access to domain-adapted solutions. Meanwhile, the rapid increase in scholarly publications across disciplines has made it increasingly difficult to stay current, further underscoring the need for scalable, AI-enabled approaches to structuring and synthesizing scholarly knowledge. Various research communities have begun addressing this challenge independently, developing tools and frameworks aimed at building reliable, dynamic, and queryable scholarly knowledge bases. However, limited interaction across these communities has hindered the exchange of methods, models, and best practices, slowing progress toward more integrated solutions. This manuscript identifies ways to foster cross-disciplinary dialogue, identify shared challenges, categorize new collaboration and shape future research directions in scholarly knowledge and organization.
The presentation of the Interdisciplinary Colloquium on Digitalisation of Research "Transformation of Medical Care and Research Based on Digitalisation - Health 4.0" with Till Keller.
- Challenges and requirements of modern health care and medical research
- Real world data, pitfalls and chances
- Artificial intelligence as aide in daily clinical routine
- Machine learning as opportunity in medical research
The presentation of the kick-off meeting of the Interdisciplinary Colloquium on Digitalisation of Research "Is Your Research Ready for the Digital Revoution?" with Sonja Schimmler, Anna Jacyszyn, and Felix Bach.
Dieses Fact-Sheet entstand aus einem Fragenkatalog des Use Cases „Publikation großer Datensätze“ des DiTraRe-Projekts. Es soll als Übersicht und Ressourcensammlung für in der Forschung häufig auftretenden Fragen dienen. Die Fragen haben ihre rechtlichen Schwerpunkte im Urheber- und allgemeinem Datenrecht.
This DiTraRe Tandem Talk by FIZ Karlsruhe and ITAS at EScienceTage 2025 explores the transformative impact of digitalisation and artificial intelligence (AI) on scientific workflows and public trust in research. Introducing the Leibniz Science Campus project DiTraRe, the talk focuses on two key dimensions—Reflection & Resonance and Tools & Processes—and examines a practical use case, Smart Labs, through the Chemotion Electronic Lab Notebook. Highlighting how structured digital workflows improve reproducibility, data transparency, and FAIR data management, the presentation discusses AI's roles as a computational microscope, source of inspiration, and potential autonomous agent of understanding. It also addresses challenges around research integrity, AI explainability, and regulatory complexities. Attendees are invited to engage further via the symposium, social media, and the project's website https://ditrare.de.
This presentation was given at the E-Science-Tage 2025 at the University of Heidelberg. It highlighted the potential technical and legal challenges, as well as the benefits, of using electrophysiological models, known as digital twins in biomedical engineering, to generate synthetic data such as electrocardiograms.
This is a presentation of the workshop "Information Infrastructures in the Era of AI: Opportunities and Challenges" organised by the DiTraRe team which took place on the 12th of March 2025 during the E-Science-Tage in Heidelberg.
This poster presents the Leibniz Science Campus Digital Transformation of Research (DiTraRe) in a concise way.
This talk was presented as a Lightning Talk during the E-Science-Tage 2025 in Heidelberg.
https://zenodo.org/records/14990383
This is a ready-to-publish version of the interim report submitted on the 14th of February 2025 by the Leibniz Science Campus "Digital Transformation of Research" (DiTraRe) to the Leibniz Association. The report summarises the first 16 months of activity of DiTraRe.
In the petabyte-era of climate research, harmonising diverse environmental and geoscientific datasets is critical to improve data interoperability and support effectiveness of interdisciplinary studies. This paper presents an idea of designing an LLM-based tool to extract and standardize metadata from climate research repositories. The solution leverages the adaptability of LLMs that are able to understand contextual nuances. By addressing common inconsistencies such as varying parameters (observation types), units, and definitions, the proposed tool will significantly improve effective data integration. It will be the first step to facilitate the creation of a unified metadata schema adhering to the FAIR principles.
https://zenodo.org/records/14925185
In the petabyte-era of climate research, harmonising diverse environmental and geoscientific datasets is critical to improve data interoperability and support effectiveness of interdisciplinary studies. This paper presents an idea of designing an LLM-based tool to extract and standardize metadata from climate research repositories. The solution leverages the adaptability of LLMs that are able to understand contextual nuances. By addressing common inconsistencies such as varying parameters (observation types), units, and definitions, the proposed tool will significantly improve effective data integration. It will be the first step to facilitate the creation of a unified metadata schema adhering to the FAIR principles.
Digitale Transformationsprozesse wirken sich auf die Forschung & Wissenschaft aus. Einerseits bei der Erhebung, Auswertung & Aufbewahrung von Forschungsdaten, andererseits beim Einsatz digitaler Technologien auf die Forschenden selbst sowie auf die Gesellschaft. Eine besondere Rolle spielen dabei generative KI-Anwendungen, beispielsweise für die Recherche oder Analyse von Daten. Vor diesem Hintergrund ergeben sich auch neue Fragestellungen für die TA, etwa die Politikberatung. Im Projekt „Leibniz WissenschaftsCampus – Digital Transformation of Research“ (DiTraRe) widmen wir uns diesen Themen.
The recently established Leibniz Science Campus “Digital Transformation of Research” (DiTraRe) investigates the effects of a broadly understood process of digitalisation of research on a multilevel scale. The project concentrates on four research clusters concerning different topics and gathering use cases from varying scientific areas. For a multi-scale investigation these research clusters are interwoven with four dimensions, each of which approaches the tasks from a different perspective and poses its own research questions. Within this “spider’s web” we are not only developing practical solutions for each use case but also seeking to find generalisations valuable to the scientific community as well as society in general. Sophisticated AI technologies, like natural language processing, knowledge extraction, and ontology engineering, are investigated within the DiTraRe project by the dimension Exploration and knowledge organisation. This position paper aims to describe the DiTraRe Science Campus in general as well as concentrate on its aforementioned dimension concerning implementation of AI techniques.
This vision/position paper shortly introduces and summarizes the DiTraRe project with a special focus on implementation of AI techniques to each use case by the dimension "exploration and knowledge organisation". The paper was accepted to be published and presented during the 4th International Workshop on Scientific Knowledge: Representation, Discovery, and Assessment, co-located with the 23rd International Semantic Web Conference (ISWC) 2024.
In our contribution to the fPET 2024 conference (Forum on Philosphy, Engineering and Technology 2024), we present the concept of a Leibniz ScienceCampus and the specific characteristics of DiTraRe. One focus is the interface between natural sciences and engineering on the one hand and sociology, technology assessment, law and ethics on the other. We report on our experiences with the chosen interdisciplinary and multi-layered approach and present the first results of our work with a focus on current challenges.
A poster presenting the DiTraRe project in general as well as focusing on the current state and ideas of applying specific AI methods to each use case by the dimension "exploration and knowledge organisation". The poster was presented during the fPET 2024 conference (Forum on Philosphy, Engineering and Technology 2024).
The effective refractory period (ERP) is one of the main electrophysiological properties governing arrhythmia, yet ERP personalization is rarely performed when creating patient-specific computer models of the atria to inform clinical decision-making. This study evaluates the impact of integrating clinical ERP measurements into personalized in silico models on arrhythmia vulnerability.
Computer models for simulating cardiac electrophysiology are valuable tools for research and clinical applications. Traditional reaction-diffusion (RD) models used for these purposes are computationally expensive. While eikonal models offer a faster alternative, they are not well-suited to study cardiac arrhythmias driven by reentrant activity. The present work extends the diffusion-reaction eikonal alternant model (DREAM), incorporating conduction velocity (CV) restitution for simulating complex cardiac arrhythmias.
The role of the right atrium (RA) in atrial fibrillation (AF) has long been overlooked. Multiple studies have examined clinical conditions associated with AF, such as atrial enlargement, fibrosis extent, electrical remodeling, and wall thickening, but have been mainly concentrated on the left atrium (LA).
Computer models of the atria can aid in assessing how the RA influences arrhythmia vulnerability and in studying the role of RA drivers in the induction of AF, both aspects difficult to assess clinically and experimentally. This work assesses the “Creative Concept” of incorporating the RA in computational arrhythmia studies based on 1398 virtual pacing sequences in 8 biatrial and 8 monoatrial patient-specific models under 3 different substrate conditions, resulting in a total of 48 distinct model configurations.
This is the original proposal for the Leibniz Science Campus "Digital Transformation of Research" (DiTraRe). Within DiTraRe, the effects and potentials of the increasing digitalization of scientific work are investigated in four interdisciplinary research clusters. Based on use cases from different disciplines, concrete solutions for challenges arising from digitalisation are developed, which are then generalised.
A poster introducing the DiTraRe project to the community working on applying AI techniques in research presented at the workshop AI4RE (AI for the Reserach Ecosystem). A short wrap-up of the project in general together with a more detailed yet still preliminary view from the perspective of the dimension "exploration and knowledge organisation".