Chemistry Data Days 2025

Second edition of a success story

On 3rd and 4th June 2025, the Chemistry Data Days took place for the second time at Johannes Gutenberg University Mainz, attracting almost 80 participants to the Atrium Maximum.

The first day focussed on the question of how to teach good research data management in a curricular environment to students of chemistry and related subjects.

Georg Manolikakes (RPTU Kaiserslautern), Sonja Herres-Pawlis (RWTH Aachen University), Heiko Weber (FAIRMat, FAU Erlangen), Ingo Mey (Georg-August University Göttingen), and Albert Engstfeld (Ulm University) reported on how they have already integrated an electronic lab notebook (ELN), Jupyter or basic research data management into their teaching. There were different approaches, but where student surveys were available, the results were always clearly positive. The handling of data in line with the FAIR principles is becoming more and more natural and students are happy to take up the digital literacy programmes where they are offered. Therefore, several docents are planning to publish their approach in the near future, which will provide a deeper insight.

Nicole Parks, Data Steward of the SFB 985 at RWTH Aachen University, then provided insights from the perspective of those who support researchers in data management. Many participants agreed with her point that data organisation benefits the researcher and their working group first, long before the data is published.

In the early evening, Till Biskup posed the question of the relationship between data (publication) and good science, before the first day came to an end with the poster session and a buffet. A poster prize provided by the journal Digital Discovery (RSC) went to Bernadette Mohr from HU Berlin, member of the FAIRMat consortium.

The second day focussed on data tools supporting the data lifecycle. Presentations on our data management plan (DMP) templates, the Chemotion ELN and the selection of the most suitable repository provided insights into the wide range of services offered by our consortium.

The focus then shifted to automation and AI. Felix Bänsch and Markus Nietfeld (Beilstein Institute) presented BChemXtract, a new innovation for all Beilstein-Institut journals that will soon be launched. It automatically converts all molecular data contained in publications into a machine-readable form, significantly improving the reusability of data. Mathias Christmann presented an AI-supported tool that automatically checks the accuracy of mass data in publications after finding out in a study that only 40% of all publications tested contained correct mass data. Finally, Sandra Geisler showed how LLM can already provide useful support for the organisation and processing of data.

The lively discussion on AI-supported tools at the end showed where research data is heading: automation will make many steps from data collection to publication much easier, although there is still a lot of work to be done.

Once again this year, the CDD could not have taken place without the support of our partners and sponsors, to whom we would like to express our special thanks: