4. Workshop: Ontologies4Chem
November 11 , 9:00 – November 13 , 18:00 CET
Final Agenda
We are pleased to share the final agenda for the 4th Ontologies4Chem Workshop 2025 in Limburg an der Lahn (Hesse, Germany, 0,5 hour by train from Frankfurt Airport) which is jointly organized by NFDI4Chem, NFDI4Cat, Beilstein-Institut & PSDI.
Registration for on-site participation is closed due to capacity limits being reached. However, we are offering all interested parties the opportunity to participate online, as we will be live-streaming the main talks scheduled in the pre-noon sessions (9:00 a.m. to 12:00 p.m. each day). For more details, including short abstracts of the planned talks, please refer to the agenda.
To register for online participation, please email the Workshop Organization Team at ontologies@nfdi4chem.de.
Workshop Goals
A key aim of Ontologies4Chem is to agree on a “canon” of recommended chemistry-related ontologies (meaning a set of widely accepted, interoperable ontologies) for metadata annotation of FAIR research data.
We plan to improve the documentation and governance of these ontologies directly within the NFDI4Chem Terminology Service (TS). The TS, which is built upon EBI’s OLS, is our intended „single point of truth“ for documenting the canon. We’ll leverage its unique features, such as: integrated issue filing with underlying ontology repositories; and the user-specific note-taking functionality, to capture crucial context (which projects use which ontologies, how specific terms are utilized, and what gaps exist).
To effectively vet and improve this canon, we need real-world examples and practical feedback.
Your Contribution: Data or Context
We invite you to contribute to the hands-on „hacking“ sessions by sharing your experiences in one of the following two ways:
1. Data Examples for Gap Analysis
Please bring an example of data that you are either planning to semantify with chemical ontologies or data that you have already annotated.
The intention is to use these concrete examples to stress-test existing chemical ontologies and immediately identify where they currently have gaps, ambiguities, or missing terms. This input will be directly used to file issues and add context notes within the NFDI4Chem Terminology Service, ensuring our documentation is project-relevant and current.
(Note: If the data belongs to a colleague, please obtain their permission and approval beforehand, and ensure that you are complying with the usage terms of the data.)
2. Ontology Usage Contexts, Barriers, and Existing Issues
If you don’t have a concrete data example ready, we are still keenly interested in your general experience and challenges. Please be prepared to discuss, or ideally, bring:
- Existing Unresolved Issues: We strongly encourage you to bring forward an existing, open issue from the issue tracker (e.g., GitHub repository) of a chemical ontology. These sessions are planned to help us collaboratively tackle and resolve these existing tickets.
- Usage Context and General Barriers:
- Which chemical ontologies you currently use and in what context.
- The specific issues or barriers you encounter when trying to use existing ontologies for annotation.
- Ontologies Not Yet Indexed: Since our TS index is not yet exhaustive, we welcome proposals for other chemistry-related ontologies that should be added to our service to achieve a more common canon.
Your practical input is essential for creating a reliable and actionable canon and documenting its usage within the TS. We look forward to a productive and collaborative workshop!
Agenda: Overview & Detailed Information
The core program of the workshop will be held from Tue, Nov 11 to Thu, Nov 13, with a “Welcome Dinner” on Monday evening (Nov 10) and departure on Friday morning (Nov 14) after breakfast. Each day will feature morning presentations (live-stream available) to inspire engaging discussions and hands-on hacking sessions in the afternoon.
| Time/Day | Mon, Nov 10 | Tue, Nov 11 | Wed, Nov 12 | Thu, Nov 13 | Fri, Nov 14 |
|---|---|---|---|---|---|
| 08:00 – 09:00 | Breakfast | ||||
| 09:00 – 10:00 | Welcome & Opening Keynote | Talks around the topic “Ontological Representation of Chemical Reactions” | Talks around the Topic “Ontological Representation of Chemical Entities” | Departure | |
| 10:00 – 12:00 | Talks from NFDI consortia, PSDI etc. | ||||
| 12:00 – 13:00 | Lunch | ||||
| 13:00 – 16:00 | Arrival & Registration | Updates from Ontology Projects & Applications Build Upon Ontologies | Hacking Sessions | Hacking Sessions | |
| Preparation for the Final Round | |||||
| 16:00 – 18:00 | Hacking Sessions Ontology Curation & Tools | Social Event | Final Round (Reporting back on Hacking Session / Final Discussion) & Closing of the Workshop | ||
| 19:30 | Dinner | ||||
DAY 1 (Tue, Nov 11): Chemistry Related Projects & Ontology Development
Focused on projects developing, curating, or implementing ontologies within a FAIR data management context.
| 09:00 – 10:00 | Welcome & Opening Keynote |
| >>> | Talks from NFDI consortia, PSDI etc. |
| 10:00 – 10:30 | How are we standardising metadata in the Physical Sciences Data Infrastructure (PSDI)?The first version of the Physical Sciences Data Infrastructure (PSDI) was launched in spring this year. Here we describe the metadata profiles that were included in PSDI, and the standards used in those and give an example of how these are used to plug a resource into PSDI and its services. We also outline areas for future development. |
| Coffee break (15 min) | |
| 10:45 – 10:55 | Updates from the MADICES (Machine-Actionable Data Interoperability for the Chemical Sciences) workshopIn late October 2025, the MADICES 3 workshop brought together a diverse international community of researchers, developers, and practitioners to explore the complexities of interoperability across different aspects of the chemical sciences. This presentation will share key findings from the event, highlighting the discussions and practical activities that were undertaken to address these challenges, and reflect on the progress that has been made over the past three events. |
| 10:55 – 11:05 | Ontologies4Chem Status UpdateWe will present the progress we’ve made in curating and developing the chemical ontologies used in NFDI4Chem, such as CHMO, nmrCV and VIBSO. |
| 11:05 – 11:15 | Towards an Ontology for Theoretical ChemistrySurprisingly, no comprehensive ontology exists for theoretical chemistry. Several partly overlapping projects already address fragments of theoretical chemistry or focus on specific software packages (e.g. OntoCompChem or MolSSI QCSchema). In the process of building a comprehensive ontology for theoretical chemistry we should align with these existing vocabularies to maximise interoperability. Early adoption can be catalysed by embedding identifiers into widely used computational packages (e.g., ORCA) and data repositories (e.g., Chemotion, NOMAD, QCArchive), thus ensuring immediate practical value. By uniting theoretical chemists around a shared semantic framework, the proposed ontology will accelerate method innovation, improve software sustainability, and unlock new possibilities for machine learning. |
| 11:15 – 11:30 | Ontology-based recoverability assessment of critical raw materialsIn this contribution, we present the design and integration of two complementary ontologies to support the recoverability assessment of critical raw materials. The first ontology formalizes composition data of end-of-life vehicles using the dataset established within the FutuRaM Horizon Europe project, capturing hierarchical relationships across product–component–material–element compositional levels. The second ontology encodes the “metal wheel of recycling”. The ontologies are connected through ChEBI (Chemical Entities of Biological Interest) identifiers, which align the elements across both representations. The resulting knowledge representation enables seamless navigation across composition levels and provides actionable guidance for identifying the recovery possibilities for selected elements and materials. This integrated, extensible framework improves traceability and supports data-driven decision making in recycling. |
| 11:30 – 12:00 | Limits and solutions for ontological representation of chemical reactions with LARAsuiteThis talk will present the LARAsuite, a freely and openly available collection of applications, libraries, databases and tools to plan, manage, create, monitor and evaluate (automated) processes in the laboratory. The focus of this talk will be on how chemical ontologies can be used to annotate reaction data to produce ontology-grounded metadata necessary for publishing reaction data as FAIR research data. |
Lunch Break (60 min) | |
| >>> | Updates from Ontology Projects & Applications Built Upon Ontologies |
| 13:00 – 13:30 | Extending DCAT-AP for chemistry & catalysisWe will present the DCAT-AP extension developed in a joint venture by NFDI4Chem & NFDI4Cat. This LinkML based, extended version of DCAT-AP can be used to provide domain specific metadata on datasets. We will demonstrate how this extension can be applied for any scientific domain due to its mapping to PROV-O, DCTerms and QUDT, as well as how it can be further extended easily to provide domain specific metadata schemata directly mapped to more domain specific terminologies on the example of chemistry and catalysis. |
| 13:30 – 13:45 | Terminology Services 4 NFDI (TS4NFDI)TS4NFDI is a cross-domain service for the provision, curation, development, harmonization, and mapping of terminologies. It aims to facilitate consensus-building and interoperability of services across disciplines to achieve a shared knowledge representation and knowledge engineering framework. The purpose of this presentation is to provide a comprehensive overview of the current development status of TS4NFDI, with particular emphasis on its interactions with other systems. |
| Coffee break (15 min) | |
| 14:00 – 14:30 | Implementing Webby FDOs with RO-Crate and FAIR SignpostingRO-Crate offers a streamlined way to bundle research outputs with their metadata, FAIR Signposting enables intuitive navigation of scholarly resources online. Together, these technologies create a web-friendly or, in short, webby implementation path for FAIR Digital Objects. The approach can be easily integrated into existing data infrastructures or used for ad-hoc research objects on standard web hosting platforms. |
| 14:30 – 15:00 | Benefits of using RO-Crates to share Galaxy supported analysisRO-Crates facilitate publishing of FAIR data. In this talk we discuss how they are integrated into the Galaxy environment, the main benefits from this integration and the work we are doing to complement them and make them more relevant for the wider research community. This includes talking about reproducibility, replicability, as well as extending findability using domain metadata. |
| 15:00 – 15:30 | ARC²: Annotated Research Context & Annotated RO-CrateThe ARC² framework (Annotated Research Context and Annotated RO-Crate) introduces a declarative, provenance-driven approach to describing research data through the processes that generate them. ARC combines a lightweight file- and folder-based scaffold for organizing data and metadata with a standards-compliant RO-Crate representation for machine-actionable publication. Building on the ISA model and Common Workflow Language, it unifies experimental and computational provenance in a consistent, semantically rich format. Through its modular design and supporting tools, ARC makes structured, FAIR, and interoperable data description accessible to researchers across disciplines. |
| Coffee break (15 min) | |
| 15:45 – 16:00 | Hacking Session introduction / planning |
| 16:00 – 18:00 | Hacking Sessions „Ontology Curation & Tools“ |
DAY 2 (Wed, Nov 12): Representing Chemical Reactions with Ontologies
Focused on continuing the discussion around which ontologies can and should be used to represent chemical reactions.
| >>> | Talks around the topic „Ontological Representation of Chemical Reactions„ |
| 09:00 – 09:30 | Ontology pre-training improves prediction of molecular propertiesWe are presenting how we use ontologies to improve Machine Learning models using a method we call “Ontology Pre-Training” (https://www.biorxiv.org/content/10.1101/2025.09.30.679573v1). We transform the ChEBI hierarchy into a classification problem and add it as an additional training step of a chemical foundation model. We show that this method improves performance for many downstream applications of small molecules, such as toxicity, aqueous solubility, and lipophilicity prediction. |
| 09:30 – 10:00 | Ontological representation of experiments in electrochemistry and synthetic chemistryThis talk will present a high-level abstraction model designed for the ontological representation of chemical experiments. Using two distinct and complex use cases, high-throughput electrochemical testing and automated organic synthesis, the presenters will demonstrate how this model can capture diverse experimental scenarios. The primary goal is to engage the audience in a discussion to identify and address common challenges, as well as to pinpoint any gaps or weaknesses in the model, laying the groundwork for the afternoon’s hacking session. |
| Coffee break (30 min) | |
| 10:30 – 11:00 | RDF model of the Rhea reaction databaseRhea is an expert-curated knowledge base of reactions that is used in UniProtKB |
| 11:00 – 11:30 | React4Cat – classifying reaction via functional groups and SHACL-AFClassifying datasets based on their content is an important task for making data FAIR. In the domains of chemistry and, more specifically, the subdomain of catalysis, this is most often not a trivial task, since experiments usually lead to side reactions, which should also be documented. This is where an optimized Reac4Cat aims to ease this task by classifying datasets against established reaction databases and by comparing and classifying reactions based on functional group chemistry. To facilitate this, classic ontology logic is not sufficient. Therefore, an additional classifying mechanism based on SHACL’s advanced feature set is implemented, using SAPRQL-based construct queries to establish new relations and add an extra layer of logic. |
| 11:30 – 12:00 | Hacking Session introduction / planning |
Lunch Break (60 min) | |
| 13:00 – 16:00 | Hacking Sessions „Ontological Representation of Chemical Reactions„ |
| 16:00 – 18:00 | Social Event |
DAY 3 (Thu, Nov 13): Representing Chemical Entities with Ontologies
Focused on continuing the discussion around the current ontologies used to represent chemical entities, especially how to reuse and improve ChEBI using AI assisted classification, design patterns, and community curation approaches.
| >>> | Talks around the Topic “Ontological Representation of Chemical Entities” |
| 09:00 – 09:30 | ChemLog: Making monadic second-order logic viable for ontological classification and learningIn this talk, we present ChemLog, a methodology that allows to use monadic second-order formalisation for ontology classification. As a case study, ChemLog has been applied to 67 peptide-related classes (including 14 current ChEBI classes). For these classes, both natural language and formalised definitions have been developed which are then used to classify both ChEBI as well as PubChem with 119 million molecules. |
| 09:30 – 10:00 | Chebifier 2: An ensemble for automated chemical classification into ChEBIChemical classification is a complex task without a one-size-fits-all solution. Rule-based approaches are explainable and reliable, but require expert knowledge. Deep Learning methods are easier to scale, but depend on the presence of sufficient data and are black boxes to the user. Chebifier 2 is an ensemble approach that combines deep learning, rule-based and LLM-based approaches to leverage their individual strengths, covering a total of more than 1,700 ChEBI classes. |
| 10:00 – 10:15 | Machine Learning bacterial biosynthesis using chemical ontologies I would like to present the work I did on my method CHAMOIS (preprint on bioRxiv) where I try to predict the structure of secondary metabolites from the sequence of their biosynthetic gene clusters. For this purpose I used the ChemOnt ontology (from ClassyFire), after having tried (unsuccessfully) with ChEBI. I would like to use this project as an example where the current top-down structure of ChEBI may not be desirable. |
| Coffee break (15 min) | |
| 10:30 – 11:00 | CHEMROF: A semantic schema for chemical entities, mixtures, and reactionsMost of the ontologies used in the biosciences are large, lightly axiomatized hierarchical terminologies, often consisting of tens of thousands of terms. The developers of these ontologies have learned the hard way that ensuring consistency of classification and connectedness is a huge challenge, unless consistent design patterns are applied. This is as true of ChEBI as it is with many other similar ontologies, and with the rollout of the ChEBI 2.0 infrastructure, now is the time to be thinking about applying higher level metaclass organization and design patterns. The CHEMROF data model was designed as both a semantic schema for chemical entities and as a metaclass-level schema for ontologies such as CHEBI, providing organization units for entities such as atoms, polyatomic entities, macromolecules, salts, racemic mixtures, both at the structural level and at the grouping class level. CHEMROF also includes classes for modeling reactions, including biochemical reactions of the kind found in RHEA. In this presentation I will describe the overarching structure of CHEMROF, how it can be used as an exchange format for chemical knowledge, and how it can help with the future evolution of ChEBI. I will also describe how we are applying the ai4curation agents to assist with the development of CHEMROF and related LinkML schemas. |
| 11:00 – 11:30 | New ChEBI Submission Portal: A step towards community curationThe modern rewrite of ChEBI, “ChEBI 2.0”, became the production version in early Oct 2025. Among other features, this encompasses a completely new submission portal. I describe ongoing efforts towards a portal to enable community curation. An important aspect of this will be a robust process to validate ontology relations. |
| 11:30 – 12:00 | Where next for ChEBI? Problems and opportunitiesThe ChEBI user meeting in Nov 2024 provided an opportunity for the community to discuss the future development of ChEBI. Here I distill some of the suggestions from this meeting, highlighting particular pain points, and discuss how these might be addressed. In particular, I reflect on the problem of having multiple alternate forms (e.g. associated with different pH), the downstream effect on users and what a potential solution might look like. |
Lunch Break (60 min) | |
| 13:00 – 15:30 | Hacking Sessions „Ontological Representation of Chemical Entities„ |
| 15:30 – 16:00 | Preparation for the Final Round |
| 16:00 – 18:00 | Final Round (Reporting back on Hacking Session / Final Discussion) & Closing of the Workshop |
Accomodation & Contact
The Dom Hotel Limburg is offering room contingents for all registered workshop participants at a price of 125 EUR per person and night (incl. breakfast). More detailed information on room reservations at the conference hotel is provided in the confirmation email after registration.
Additional hotel recommendations:
- Frankfurter Hof (right around the corner from the event hotel) http://www.frankfurter-hof-limburg.de/
- Nassauer Hof (ca. 10 min walk from the event hotel) http://www.nassauerhof-limburg.de/
- Weinhaus Schultes (ca. 10 min walk from the event hotel, located in the old town) https://schultes-weinhaus.de/
If you have any questions, please feel free to contact us via ontologies@nfdi4chem.de.
We look forward to welcoming you to Limburg!
Best regards,
The Workshop Organization Team
(NFDI4Chem, NFDI4Cat, Beilstein-Institut & PSDI)
