The 6th edition of the FAIRmat newsletter is now available for download! Stay up to date with the latest project developments, enjoy an insightful interview with NOMAD user Lena Mittmann, and explore more exciting articles from the FAIRmat community. Download it now from our website!

Due to the great public interest in the Fourth FAIRmat Users Meeting at the FAU in Erlangen, we are excited to share the recordings of users' talks and invited speakers' presentations on the FAIRmat and NOMAD YouTube channel! The full playlist includes:

• Research Data Management in Collaborative Research Centres by Brit Redöhl
• Designing RDM in collaborative funding schemes by Heiko Weber
• Knowledge management and online AI training in NOMAD by Ta-Shun Chou
• Advancing Catalysis Research by Julia Schumann
• Low-scaling algorithms developed in NOMAD by Antonio Delesma Diaz
• Taylored RDM with NOMAD by Lauri Himanen
• Employing NOMAD CAMELS in an atom beam experiment by Carina Kanitz

You can read more about the event and see the contributions here.

We're happy to announce that our tutorial on structured data extraction with large language models (LLMs) has been published in Chemical Society Reviews!

The publication, titled 'From Text to Insight: Large Language Models for Chemical Data Extraction,' provides an in-depth overview of LLM-based structured data extraction in materials science and chemistry, synthesizing current knowledge and exploring future directions. It is a product of the FAIRmat AI Toolkit task led by Kevin Maik Jablonka, with contributions from Mara Schilling-Wilhelmi (Friedrich Schiller University Jena), Martiño Ríos-García and María Victoria Gil (CSIC, Oviedo), Sherjeel Shabih, Christoph T. Koch, and José A. Márquez (Humboldt-Universität zu Berlin), and Santiago Miret (Intel Labs).

To improve interactivity and applicability, we've created a Jupyter book with practical examples tailored to chemistry and materials science. These examples can now be run directly using the Jupyter4NFDI service provided by Base4NFDI - Basic Services for the NFDI.

• Read the online book.
• Run it in Jupyter4NFDI.

The annual DGKK + DEMBE workshop took place in Berlin on December 11-13, hosted by the Paul-Drude-Institut für Festkörperelektronik (PDI) and the Fraunhofer Heinrich-Hertz-Institut (HHI). This event combined the expertise of the Deutsche Gesellschaft für Kristallwachstum und Kristallzüchtung (DGKK), focusing on the epitaxy of III-V semiconductors, with the German MBE workshop (DEMBE), bringing together leading experts and researchers in crystal growth and thin film science.

NOMAD was represented by Andrea Albino and Sebastian Brückner (Area A: Synthesis), who showcased their advancements in Research Data Management (RDM) tailored to the needs of the communities addressed in the workshop. On the opening day, they presented a poster highlighting implemented use cases for Metal-Organic Vapor Phase Epitaxy (MOVPE) and Molecular Beam Epitaxy (MBE), as well as a poster detailing the approach of FAIRmat Area A and its community plugins, which bridge the gap in adopting the FAIR principles in RDM. Additionally, Sebastian Brückner participated in a panel discussion alongside experts from academia and industry, offering insights on improving RDM practices in experimental science. The discussion addressed existing bottlenecks and explored new avenues for development, including AI-based real-time analysis for monitoring and prompt corrections during the growth process.

Throughout the event, attendees engaged with Andrea and Sebastian at the FAIRmat booth in the exhibition area. Here, they explored how FAIRmat and NOMAD tools and resources streamline research data workflows, ensuring data are Findable, Accessible, Interoperable, and Reusable (FAIR). These solutions are specifically designed to meet the needs of researchers in thin film science and crystal growth.

A significant point of interest was the reusability of the tools and data structures developed so far, which enable new use cases with minimal modeling effort. This approach strengthens shared standards within the community, fostering collaboration and ease of adoption for new users.

The workshop also attracted the attention of technology partners providing instruments for this research field. These partners showed strong interest in the RDM solutions offered by FAIRmat and engaged in productive discussions about the needs of the community and their potential role in advancing RDM practices.

On November 27, 2024, the FAIRmat community met in Berlin for the Fifth FAIRmat Users Meeting. The event focused on community building, integrating all stakeholders, and fostering collaboration with the FAIRmat consortium. It brought together researchers, data management experts, and our FAIRmat domain experts to discuss the advancements in research data management and our NOMAD tool.

During the morning session, the Users Meeting featured several invited talks by our users and collaborators, showcasing inspiring and practical applications of NOMAD, as well as collaborative efforts alongside tour sibling consortia NFDI4Cat and NFDI-MatWerk. Invited talks have been given by:

• Eva Unger (Helmholtz-Zentrum Berlin and Humboldt-Universität zu Berlin) – „The FAIRification of PV research data“
• Lena Mittmann (Technical University of Denmark) - “Enabling high-throughput materials discovery of phosphosulfides by developments in FAIR data management, visualization, and analysis in NOMAD“
• Tristan Bereau (Universtät Heidelberg) – “Martignac: Computational workflows for reproducible, traceable, and composable coarse-grained Martini simulations”
• Abril Azocar Guzman (NFDI-MatWerk) – „Achieving semantic interoperability in materials science data and simulation workflows“
• David Linke (NFDI4Cat) – “An overview of NFDI4Cat services and tools with a special focus on Voc4Cat the shared vocabularies for catalysis and related disciplines”

In the following weeks, the recordings of all invited talks will be available on our FAIRmat and NOMAD YouTube channel.

The afternoon program included two interactive workshops and a dedicated Users Help Desk. The workshops were designed to provide hands-on guidance on our NOMAD tool, aimed to empower participants to start and work with NOMAD and address user-specific challenges.

The Fifth FAIRmat Users Meeting showcased the growing community committed to research data management according to the FAIR data principles. We are thankful to all participants for their active engagement and contributions. We look forward to further collaborations and are available for support and discussion beyond the Users Meeting. Join our Discord channel to stay connected!

We are excited to see you all again at the 2025 edition of the Users Meeting.

Recordings of FAIRmat Tutorial 15 are now available on the FAIRmat and NOMAD YouTube channel! The full playlist includes:

• FAIR data management
• NeXus data modeling
• Data conversion and verification using pynxtools
• Data in NOMAD

You can read more about the FAIRmat hands-on tutorial series here. The event page has all the learning material for Tutorial 15.

Last week, our co-spokesperson Heiko Weber gave a plenary talk titled "Experimental Research Data in Materials Science and Solid-State Physics: Challenges, Strategies and Solutions" at the 8th Asian Materials Data Symposium (AMDS2024) in China.
In his talk, Heiko introduced the FAIRmat project and showed how NOMAD facilitates the management of experimental research data. The talk was attended by 1000 attendees in person and 4000 online!

We are thrilled to share our vision for advancing FAIR research data management and continuing to push the boundaries in our field.

The 2024 EOSC Symposium marked an important milestone in establishing the European Open Science Cloud (EOSC) Federation. A key highlight was the launch of the first EOSC EU Node, which will serve as a model for future nodes. Notable participants included European officials and representatives from Germany's National Research Data Infrastructure (NFDI), who emphasized the importance of alignment and collaboration among European and national infrastructures. The event drew over 450 attendees in Berlin, demonstrating its significance in promoting open science across Europe. Claudia Draxl, the spokesperson for FAIRmat, delivered the scientific keynote address titled "FAIR, what else?" which is now available online.

On November 7, 2024, we launched the third FAIRmat podcast episode!

In this series, called Pioneers in electronic-structure theory, our FAIRmat host, Miguel Marques, talks to personalities who were instrumental in developing methods and codes that have come to be known as electronic-structure theory. This work includes pioneering density-functional theory and methods beyond to treat many-body effects in ground-state and excitations.

In the third episode, Miguel Marques interviews Risto Nieminen, a former professor of theoretical physics in Finland. In the first part of the interview, Risto Nieminen shares insights into his student life, describing his experiences studying physics from 1968 on and discussing the (social) dynamics within the scientific community then. The second part of the conversation centers on his scientific achievements in density functional theory. The interview concludes with his contributions to the scientific community, including his work with the supercomputer at the CSC, his involvement in the Psi-k network, and the Millennium Prize.

Find the podcast on YouTube or enjoy the audio-only version on Spotify now!

The FAIR-DI European Conference on Data Intelligence 2024 took place from October 27 to 30 in Karlsruhe, gathering experts to discuss significant advancements in data intelligence.

The event brought together specialists from theoretical and experimental backgrounds, focusing on research data management, related services and tools, databases, and the impact of large language models (LLMs) and artificial intelligence (AI) in materials science.

Several talks introduced new databases and services for research data management during the conference, covering material synthesis, artificial photosynthesis, theoretical calculations, and polymers. Although each solution was tailored to its specific field, a common principle emerged: the importance of enabling the efficient use of substantial datasets to accelerate scientific discovery. By utilizing the potential of LLMs - such as active learning and chatbots - these tools aim to predict and discover new materials and methods while supporting scientists throughout their experiments.

A central theme of the conference was the evolving field of data intelligence. While data and intelligence are interconnected, they also influence each other independently. The community is currently focused on managing existing data in both theoretical and experimental contexts using machine learning, natural language processing, and large language models. However, these methods are expected to shape future data collection practices, raising essential questions: Which data should be prioritized? How do we ensure data quality? Can we capture all relevant data? And what role will robots play in data generation and collection? These questions were actively discussed during the conference and are anticipated to drive research in the coming years.

A highlight of the conference was the FAIR-DI Award ceremony, which recognized outstanding data handling in a PhD thesis aligned with FAIR principles. This year's award was presented to Lena Pilz from KIT. She delivered an engaging presentation on her innovative use of Chemotion as an electronic lab notebook (ELN), expanding its application to metal-organic framework (MOF) synthesis and openly publishing her research data according to FAIR principles. Congratulations!

We thank all participants of the FAIR-DI European Conference on Data Intelligence 2024 for contributing to the success of this event.