Sessions
Per starting time | Per room | Session contents
Per starting time
Per room
Alphabetical, with summary
25 years of developments in data collection, storage and distribution at the Geodesk
Maarten Storm
The Geodesk is the facility at WUR for all things related to geographic information and we are celebrating it's 25 year anniversary. This presentation will dive into our experiences in dealing with data at the Geodesk. Learn how we opened up access to geospatial data and what this meant for our processes, technical solutions and knowledge exchange.
13:55-14:40 in Room B4
A roadmap to making next-gen sequencing data FAIR
Dr. Sydney B. Wizenberg (Jordan)
Next generation sequencing is quickly becoming a popular approach for high-throughput analysis of genetic data. The pooling and processing of samples impacts data in numerous ways, all of which can influence its reusability. In this session we will outline how next-gen sequencing workflows should impact research data management practices and review a road map for making this type of data FAIR. Through a practical session on data documentation, we will identify what information is most important for reusability, and explore how the integrity of metagenomic and metagenetic applications rely on strong data management practices.
10:45-11:30 in Room B4
AI framework; standardising deep learning for reproduceability and collaboration
Tim van Daalen, Bart van Marrewijk
While working on deep learning projects, we found that much of our time was spent setting up machines, installing dependencies, and loading models—only to switch to a new one shortly after. Over the years, we explored various AI platforms and DIY solutions, both in-house and from major providers like Microsoft’s AI studio. Each attempt taught us valuable lessons, leading to the development of a new framework that ensures reproducibility while giving experts full freedom to build custom models. At the same time, it offers an easy way for other researchers to use these models. In recent months, the researchers and RITS collaborated to bring this vision to life. Join us for an inspiring session, ending with tools you can use today.
10:45-11:30 in Room A4
AI, Sensing and Automation for the next generation food sector
Aneesh Chauhan
This session will focus on how Sensing, AI and Automation technologies are coming together at Wageningen Food and Biobased Research to drive the next generation food industries.
15:00-15:45 in Room B4
An UNLOCK infrastructure for managing high-throughput timeseries data, a combined effort of two resources (LEAF & WIDE)
Jasper Koehorst
WIDE is an open-source, real-time data integration infrastructure developed with Research IT Solutions (RITS) to manage high-throughput timeseries data from diverse sources such as sensors and laboratory equipment. It comprises four modular components: data acquisition (MQTT broker), transformation (Node-RED), storage (PostgreSQL), and visualization (Grafana), with access control handled via SRAM. To connect with heterogeneous devices and protocols, WIDE receives data from LEAF, a flexible, modular system for automated data extraction from laboratory instruments. LEAF enables rapid adapter development and continuous data capture with minimal human intervention. This combined system ensures data integrity and temporal alignment, and supports the MQTT streaming standard, linking raw data to experimental context using FAIR-compliant metadata. Deployable in both lab and field environments, LEAF ensures data acquisition and WIDE enables real-time monitoring, event-driven responses, and integration with research data platforms and modelling environments.
13:00-13:45 in Room B6
Data legislation that apply to WUR
Luc Boelhouwer
t.b.d.
09:30-10:15 in Room B2
Development of centralized Research Data Management at Wageningen Marine Research
Martin Klompmaker
Over the last six years Wageningen Marine Research has established a data management support team for the research data being collected at the institute. With the aim to develop and maintain centralized data infrastructure for data collection, data storage and data analysis. Currently still in transition our strategy evolves around open-source tooling, centralized databases, and governance of data management budgets.
15:00-15:45 in Room B6
Digital Future Farm (DFF) and ethical, legal, and social AI (ELSA) aspects
Marcia Stienezen, Mireille van Hilten, Marc-Jeroen Boogaardt
Within WUR we work at the development of the ELSA scan (ELSA=ethical, legal and social aspects of AI). Within WUR we work at the development of the digital twin Digital Future Farm (DFF) and have to deal with ELSA aspects. The DFF was tested in the ELSA scan. Based upon this we would like to prepare a workshop and discuss ELSA aspects with participants of the WMDD.
10:45-11:30 in Room B2
FARM DATA: how to support on-farm digitalisation, digi-adoption and innovative research
Tamme van der Wal
More and more projects are concerned with the digitalisation at the farm and develop data infrastructures for collecting, harmonising and (re)using farm data. This workshop is to exchange ongoing initiatives, and develop a WUR wide strategy.
10:45-11:30 in Room B6
Future proof management of models on soil-water-plant interactions
Mirjam Hack, Marius Heinen, Allard de Wit, Mechteld ter Horst
This session wil focus on ideas for keeping our main models, that are frequently used for national and international policy making and scenario studies, updated and modernized. Our 'clients' expect us to manage this well, but how do we actually keep our models in shape? How do we train new researchers for this topic? What is required to make our models future proof? Let's share ideas on how to reserve time and capacity for these important tasks and how we have achieved this up to now. The models to be addressed are SWAP, WOFOST, ANIMO, (Geo-)PEARL and TOXSWA.
13:00-13:45 in Plenary room
Getting started with foundation models (large AI models): a hands-on introduction
Daan Korporaal
Foundation models are currently a very big trend within AI. In this session we will teach you exactly what they are, how they can benefit your projects, and how you should utilize them. First, we will teach you some theoretical background behind this model type and why they are currently as powerful and useful as they are. Then, we will give concrete examples of how we at Wageningen Food Safety Research use foundation models in many of our projects. Finally, we will conclude the session with a practical workshop where you can try playing around with some of these models yourself to get a more concrete idea of what they are capable of. Bring your laptop for a hands-on session!
09:30-10:15 in Room A4
GreenLight2.0 - an open platform for modelling dynamic systems
David Katzin
GreenLight2.0 (https://github.com/davkat1/GreenLight) is an open-source Python platform aimed at facilitating open science in dynamic greenhouses, with a focus on greenhouse climate and crop modelling. With this platform, users can easily transform models published in literature to a working simulation model, and then evaluate, adjust, extend, and combine models and model components. This interactive session will include: a presentation of GreenLight and what it can do; a demonstration of how GreenLight can be used; and a live workshop where users can create, modify, or combine models themselves. Bringing laptops, and publications of dynamic models that users wish to implement, is encouraged.
10:45-11:30 in Room A3
How do I publish my research data and software?
Koen Bokhorst, Ignacio Sancho
In this session, we will show you the importance of publishing your research data and software. We will provide hands-on tips to make your data and software more FAIR, and through practical examples, we will demonstrate how to increase the visibility and impact of your research. Topics covered in this session include: selecting a repository, data documentation, Git integration, and licences for data and software.
09:30-10:15 in Room B4
How to get started with (actually) making software, models or AI available for society?
Boaz van Driel, Carla Kalkhoven
At Co-creating Value and Impact (CVI), we are the team that zooms in on digital and AI cases for valorisation. We want to inspire the attendees to think about the process of getting digital inventions and solutions to society. Topics we like to cover / questions to investigate: 1) Ideation with a market / end-user in mind, 2) Developing a digital product / solution within WUR (but for use outside WUR) and 3) How can CVI support the first steps in value creation?
15:00-15:45 in Room B2
iRODS & Yoda: research data management tools
Joris Luijsterburg, Nika Valencic Kool
Amongst various Research Data Management (RDM) services offered at WUR we will dedicate this session to discuss two of them Yoda and iROD - their similarities, differences and (specific) uses. Towards the end of the session, we will split into two groups where listeners can choose to follow a short usecase/demonstrations of iRODS (lead by Joris) or ‘publishing in Yoda’ (lead by Nika).
13:55-14:40 in Room B6
KB projects AI and modelling
Jene van der Heide, KB project leads
t.b.d.
15:00-15:45 in Room A3
Language models in biology
Aurin Vos
Language models (LMs) are advancing biology by enabling sequence-based predictions across proteins, genomes, and small molecules. Protein LMs (pLMs) like ESM2 capture structural and functional patterns from evolutionary data, supporting applications from annotation to design. Genomic LMs (gLMs) such as Evo2 learn regulatory syntax and recover annotation-like features from raw DNA sequences, opening new possibilities for functional interpretations and sequence engineering. In chemistry, SMILES-based LMs and graph neural networks are driving progress in molecule generation and property prediction. LMs are beginning to reshape how we understand biological function, regulation, and interaction. This session explores their emerging role across molecular biology.
13:55-14:40 in Room A4
LTER-LIFE as a model coupling environment
Geerten Hengeveld, Nafiseh Soveizi, Stanley Nmor
LTER-LIFE is a Dutch research infrastructure to support the digital twinning of ecosystems. In this session we want to explore with the audience how they build their modelling workflows and how LTER-LIFE could be used (or should be designed) to support this process. For more information and a preview of tools we are developing, see www.lter-life.nl or www.lter-life-experience.org.
09:30-10:15 in Room A3
Model Management in Action: A serious Game
George van Voorn, Cheng Liu, Marjolein Bakker, Maarten Braakhekke, Ab Veldhuizen
In this session the goal is for the participants to jointly and actively undergo a below-an-hour serious game around computer modelling to experience both the fun and the typical challenges modellers face when doing their job. We will then explore what model management support and guidelines are in place at Wageningen University & Research. We will look at topics like: What is the Status A checklist for modellers, machine learning and databases? Who is it for? Where can I get help to ensure my modelling work is in line with the WUR standards? And why do we even bother with this time-consuming, costly model management at WUR?
09:30-10:15 in Plenary room
Modelling and observatories for sustainable Food Systems monitoring within the FutureFoods partnership
Lan van Wassenaer, Ali Hürriyetoğlu, Johannes Kruisselbrink
The FutureFoodS partnership seeks to coordinate, align, and amplify European and national research and innovation efforts to future-proof food systems through co-beneficial, integrated, and transdisciplinary approaches. By combining scientific evidence with collaborative engagement among practitioners and citizens, FutureFoodS aims to support transformative change at local, national, European, and global levels. Work Package 5 of the partnership focuses on developing a Food Systems Observatory - a platform for sharing methods, metrics, data, and assessments related to the sustainability performance of food systems. At Wageningen University & Research, we envision this observatory as a distributed network of collaborative nodes, each situated within consortium partner regions and embedded in their respective data infrastructures, research communities, and stakeholder ecosystems. This session will explore the modelling opportunities and challenges inherent in realizing this ambition, and invite discussion on how to build a robust, inclusive, and action-oriented observatory for sustainable food systems transformation.
13:55-14:40 in Room A3
Modelling, Policy and Responsibility
Joske Houtkamp, Marcel Pleijte, Rogier Pouwels, Geerten Hengeveld
Models are important tools in policymaking — but what happens when their use leads to contested outcomes or problematic decisions? Who is responsible when a model is misapplied, misunderstood, or produces results that are challenged in public or political debate? This question becomes even more complex considering that models often require scientific expertise to interpret, involve uncertainties, rest on assumptions, and may contain errors. They are sometimes applied beyond their intended scope — yet they can have a significant impact on policy decisions. This interactive workshop invites data experts, modellers, researchers and others interested to reflect on their role and responsibility in this grey area. We’ll discuss real-world dilemmas: • Are we aware of how our models will be used? • Do we feel responsible for how they are applied? • Have we experienced difficult situations around model use? Through short examples and open discussion, we will probe how responsibility is shared (or avoided) between modellers, policymakers, and others.
13:55-14:40 in Room B2
Modernising and integrating legacy models: The case of ReNEMA
Domantas Giržadas
Numerous computation models used for policy evaluation are implemented in Excel workbooks, Genstat scripts and other aging frameworks. Maintaining these legacy models is gradually becoming burdensome. Modelers are dealing with inflexible frameworks, while managers are having trouble finding people who are able (and willing) to work with “old-fashioned” systems. In the ReNEMA project, Wageningen Environmental Research (WEnR) is working together with Centraal Bureau voor de Statistiek (CBS) on reimplementing and harmonising a set of agricultural emissions models in Python programming language and combining them into one modern, accessible, and sustainable modelling platform. Along the way, the ReNEMA project faced us with challenges of model co-development with external partners, protection of sensitive data, and inter-disciplinary communication. Curious about our journey? In this presentation we will share our joys, sorrows, and insights so far. Are you dealing with similar challenges? We would love to exchange tips and ideas!
13:55-14:40 in Plenary room
Partnership in software development (from model to software)
Peter van der Togt
Within WUR, software development takes place in various departments. We have identified the following issues: there is no uniform way of working, different technologies are being used, legacy software is still in use, and knowledge is concentrated among too few people. To ensure the continuity of software development within WUR, it is essential to cooperate and to standardize the way of working, moving towards modern techniques. This would also help ensure that WUR remains attractive to young developers. In this session, we want to explore whether we can collaborate and what role WUR-IT could play in this.
10:45-11:30 in Plenary room
Personal data: obtaining consent and reviewing ethical guidelines
Privacy Officers, Research Ethical Committee
If you work with personal data in research, you must meet a number of conditions. It’s crucial to follow legal and ethical guidelines, particularly under regulations like the GDPR. What these guidelines imply might raise questions. In this session, we guide you through this process and make you more comfortable how to work with personal data in research projects.
13:00-13:45 in Room B2
Test good practices on FAIR sharing of Datasets, Soilwise-HE project needs your input
Paul van Genuchten
As SoilWise-HE project we support the Horizon Europe Soil-Mission programme on data and knowledge provisioning and reuse. As part of the initiative we are testing a range of strategy documents on how to optimally share your research outputs. We invite you to test these documents on your research deliverables, for us to understand how we can improve them. We currently have strategy documents for sharing: vocabularies/glossaries, CSV datasets, SQLite datasets, project websites/knowledge hubs. Any feedback welcomed.
13:00-13:45 in Room B4
The National Key Registry of the Subsurface: a BRO-mance about data, models and users
Tom Harkema, Marlies de Keizer, Kees Teuling, Dennis Walvoort
The National Key Registry of the Subsurface (Dutch: “Basisregistratie ondergrond”) provides data and models about soil, hydrology, landforms, and geology. Wageningen Environmental Research is responsible for three models and associated datasets. These models and datasets are potentially useful for many WUR-researchers. After a brief introduction to the National Key Registry of the Subsurface, we’ll explain how we collect our data, how we build our models, how we improve model quality and reproducibility, how we interact with end-users and how we deploy our data and models. We also give a live demonstration on how our data can be retrieved via the official data portals.
15:00-15:45 in Plenary room
Unlocking research knowledge with AI
Rutger Vlek
Shaping future food systems to be sustainable and deliver sufficient, nutritious and safe food requires access to knowledge from various scientific domains. In this session we will share results and insights from a Knowledge Base (KB) project aimed at exploring the potential and limitations of modern generative artificial intelligence (AI) for unlocking such research knowledge. This research focusses on specific use-cases, including side-stream valorization, the protein transition and personalized dietary advice, but most findings are applicable to other subject areas of WUR.
15:00-15:45 in Room A4
Unlocking the WUR Library Data lake: From Siloed Assets to Open, AI-ready insights
Valentijn Evers, Cristina Huidiu
WUR Library holds a wealth of information—from publications and grants to datasets and over the past year, we have re‑imagined our approach to serving our use cases by committing to: • Easy sharing and interoperability and avoiding vendor lock-in through open formats and standards •Integrating AI responsibly • Leveraging existing infrastructure rather than reinventing the wheel This has led us to a modular toolbox built on Databricks, Azure, GitLab, and Terraform. While this session will not teach you how to set up any of these tools, we will share our reasoning and architectural design. The workshop will center on hands‑on exploration of strategic, research intelligence use cases. Participants will explore scenarios either directly in a Databricks environment or your data /BI tool of choice. This session is for researchers, analysts, and research support staff who want to move from isolated data to shared, insight‑driven action.
09:30-10:15 in Room B6
Utilising pre-trained AI models to analyse complex analytical instrument data
Puneet Mishra
This session will teach and demonstrate the successful use of pre-trained, publicly available AI models to address complex analytical challenges across diverse applications. The primary focus will be on analyzing sensor data—such as spectra, signals, and multivariate outputs—to predict key physical and chemical parameters of agro-food samples. Through real-world case studies, we will explore how AI models can be rapidly deployed to extract meaningful insights from high-dimensional analytical instrument data, even with limited training data. Attendees will gain hands-on understanding of how to select, adapt, and apply these models effectively for their own use cases. By the end of the session, participants will be equipped with practical tools and techniques to implement pre-trained AI models in their workflows, accelerating problem-solving and improving decision-making in research, quality control, and production settings. This session bridges the gap between cutting-edge AI and real-world analytical chemistry, making advanced tools accessible and impactful.
13:00-13:45 in Room A4
WUR Bridges High-Energy Physics and Financial Market Surveillance
Joost Pennings, Phillippe Debie
The Market Surveillance Analytics Lab (MSA Lab) is a unique interdisciplinary initiative from a six-year collaboration between WUR and CERN (HighLO). The lab utilizes methods from high-energy physics, specifically CERN’s ROOT framework, enabling efficient and scalable analysis of high-frequency financial data (300 terabyte). Designed primarily for regulators and financial exchanges, MSA Lab offers a transparent yet secure platform that avoids opaque, black-box techniques. Its core functions include statistical market analysis, intuitive visualization of limit order books (LOB), and evaluation of market impacts at high frequency (down to the microsecond). By combining frequentist statistical methods, high-dimensional sparse histograms, and efficient computational approaches, the MSA Lab translates advanced academic research into practical tools for detecting market manipulation. This approach enhances surveillance capabilities and helps safeguard market integrity. At WMDD, we will showcase the capabilities of MSA Lab, highlight our longstanding CERN-WUR collaboration, and illustrate how high-energy physics can transform financial market analytics.
13:00-13:45 in Room A3
Wageningen Model & Data Day 2025
Registration website for Wageningen Model & Data Day 2025WUR Events Teamdata@wur.nl
WUR Events Teamdata@wur.nlhttps://event.wur.nl/wageningen-model-and-data-day-2025/registration
2025-10-07
2025-10-07
OfflineEventAttendanceMode
EventScheduled
Wageningen Model & Data Day 2025Wageningen Model & Data Day 20250.00EUROnlineOnly2019-01-01T00:00:00Z
Akoesticum Akoesticum Nieuwe Kazernelaan 4 - D 6711 JC E Netherlands
