ELIXIR BioHackathon 2025

The Dutch Node, ELIXIR NL, joined the meeting with strong representation across several projects. Participants collaborated on topics ranging from interoperability and metadata standards to software FAIRness, multimodal analytics, and validation services. Several of the projects were co-led by Dutch contributors. 

Selected highlights from the projects: 

A shape‑driven visual interface to integrate heterogeneous biomedical databases into knowledge graphs (Project 1)  

This project is co-lead by Tooba Abbassi-Daloii (NL) and Javier Millan Acosta (NL). The project set the foundation for a registry of RDF dataset schemas with the Python package RDFSolve to extract, integrate and analyze linked RDF datasets. It queries for or generates VoID descriptions of the graphs in a triplestore detailing the class, property and data type partitions that constitute its actual data schema. The generated schemas are then exported into several formats, including JSON-LD, LinkML and plain text tables. Prof. Gos Micklem (UK) explored multidigraphs to find the shortest paths between two nodes in networks resulting from these schema collections, and together with Dominik Martinat (CZ) the team proposed competency questions and resources to include in the collection of schemas, including those in the IDSM, WikiPathways, the Adverse Outcome Pathway Wiki RDF, and the DBCLS RDF Portal. Next, the team plans to explore how to use the schema registry as support for query builders like BioDataFuse and Intermine, as well as data validation models across resources and in the continuous integration of RDF datasets. 

Advancing the ELIXIR Maturity Model for RDM providers with Node-level implementation guidance (Project 2) 

This project is co-lead by Mijke Jetten (Health-RI, ELIXIR-NL), Marina Popleteeva (ELIXIR Luxembourg), Xènia Pérez Sitjà (ELIXIR-UK) and Ivan Mičetić (ELIXIR-Italy). Join our cute magpie for a forest walk through the ELIXIR Data Stewardship Handbook, exploring signposts (advice and tools), campfires (stories and case studies) and waypoints (maturity indicators). With this theme, and with seven onsite and five online participants from the ELIXIR RDM Community, we worked towards a beta release of this people-focused Handbook, which also includes a Maturity Model for RDM service providers. We moved many pages to improved GitHub pages and refreshed the look and feel (based on ETT). There is still more to transfer and refine, but we are very proud of the result. Want to help build the Handbook? Contact Mijke Jetten (Health-RI, ELIXIR-NL; mijke.jetten@health-ri.nl). 

Bidirectional bridge: GitHub ⇄ bio.tools (Project 5)  

This project is co-lead by Mariia Steeghs-Turchina (NL), Anna Niehues (NL) and Veit Schwämmle (DK). The team developed a bidirectional bridge between GitHub and the ELIXIR bio.tools registry to improve research software metadata and make it more FAIR. The project is joined by Magnus Palmblad (LUMC, NL) and Walter Baccinelli (Health-RI, NL). The tooling automates metadata extraction, enrichment, and integration using APIs (GitHub, bio.tools, Europe PMC) and LLMs. It creates rich bio.tools entries from GitHub content and suggests improvements to repositories via pull requests or issues, reducing manual effort and supporting compliance with Software Management Plans. This approach lowers barriers for researchers to maintain well-annotated, reusable software. Building on the progress achieved with the fantastic collaborators during BioHackathon week, the team will continue implementing mappings and deeper integration with bio.tools. 

Embedding Multimodal Data into a Vector Database to Make a Blood Atlas (Project 7) 

This project is co-led by Tess Afanasyeva (Sanquin/Netherlands Cancer Institute), alongside Marina Pominova (BE) and Cecilia Lindskog (SE), and joined by Vedran Kasalica (Health-RI, NL). The team integrated multimodal data, combining text and omics profiles of human adult immune cell populations into a unified vector embedding dataset. They further implemented an MCP server to perform cell type-specific queries across modalities. 

Metabolomics and Proteomics file format interoperability fest (Project 14) 

During this project, Denise Slenter (Maastricht University, BioSB board) participated in a project focused on a commonly used (meta)data standard for metabolomics called mzTab-M. She focused on revamping the GitHub Action, testing the stable and released validator of this file format for several upstream software libraries (writers), and developed a new GitHub Action that directly integrates source code updates. Other team members used the developed example datasets to check their applicability in downstream analysis sources (readers).  

MiCoReCa (Microbiome Community Resource Catalogue): Towards Centralized Curation and Integration (Project 16) 

This project is co-lead by Vivek Ashokan (FR), Bérénice Batut (FR) and Nikolaos Strepis (NL). The ELIXIR Microbiome Community initiated MiCoReCa to consolidate fragmented bioinformatics resources into a dynamic, open-access catalogue. Modeled after the Galaxy Codex and launched at the ELIXIR BioHackathon, it leverages standardized ontologies and community curation to index tools across platforms like bio.tools and WorkflowHub. The catalogue further enhances utility by integrating training materials from TeSS and benchmarking capabilities from OpenEBench. Ultimately, this infrastructure aims to streamline microbiome research by strictly improving resource discoverability, reproducibility, and sustainability. 

Towards a Robust Validation Service for Data and Metadata in ARC RO-Crates (Project 28) 

This project is co-lead by Matthijs Brouwer (NL), Kevin Schneider (DE) and Eli Chadwick (GB). During the BioHackathon, the team focused on four parallel tracks: validating metadata with SHACL, payload data validation using Frictionless Data specifications, developing RO-Crate Profiles for standard definitions, and implementing ISA-structured RO-Crate export in FAIRDOM-SEEK. The initiative proved highly productive, resulting in a well-defined architecture for a dual validation system addressing both metadata and data. The data validation process, particularly the conversion to Frictionless schema, is nearing completion, with considerable advancement in the MIAPPE and ISA profile definitions. Remaining tasks include finalizing the rocrate-validator extension and further developing the Profile Crate creator tool. 

Some of the participants shared their impressions of the week: 

Tess Afanasyeva: For me, the in-person interactions with participants from diverse backgrounds at the BioHackathons in 2024 and 2025 have been an incredible source of knowledge and the starting point of many lasting friendships. 

Magnus Palmblad: This was my fourth BioHackathon, and I already look forward to the next! This time I was not co-leading a project, which allowed a bit more interaction with other projects and catching up with collaborators old and new. The venue was great and the Hub staff did an absolutely amazing job running the show. Managing an event of this scale with only four people onsite is a testament to their passion and professionalism. 

Mijke Jetten: A week of hands-on collaboration with so many people is an incredible opportunity. We could never have reached this project result working alone in our offices. The ease of finding the right people and co-working across projects makes the ELIXIR BioHackathon unique. 

Nikolaos Strepis: Participating in my first BioHackathon was an impressive experience. I was pleasantly surprised by the efficiency of working alongside individuals who share common goals within an environment that fosters genuine collaboration and brainstorming. The event also excelled in balancing technical work with social activities, allowing us to build meaningful connections. I am proud of the project we developed and look forward to attending future editions. 

Denise Slenter: As always, the ELIXIR Biohackathon is a great place to work with a clear focus on a dedicated project and meet many new people! 

Javier Millan: This is my second time joining the ELIXIR Biohackathon, and I’m already looking forward to the next one. Thanks to everyone who joined our project, showed interest, and provided useful help, and especially to the organizers for their good work. These events are a great opportunity to learn from others and get a lot of hands-on work done in little time. 

Anna Niehues: The BioHackathon provided a valuable opportunity to collaborate with enthusiastic colleagues from Europe and beyond. I'm looking forward to continuing the collaborations started there. 

Mariia Steeghs-Turchina: This was my first BioHackathon, and I was really impressed by how much progress we achieved by working with a motivated and genuinely cool group of people. We worked long days and still found ourselves merging pull requests in the evenings with a glass of wine, which was quite exhausting but also a lot of fun. Can’t wait for the next edition!