Reproducible code & Coffee

How can we make our code easier to understand, rerun, and reuse — not only for our colleagues, but also for our future selves?

In this informal, researcher-led session, we explore practical ways to improve computational reproducibility through small, everyday habits. No lectures, no slides — just real examples, honest experiences, and plenty of room for questions.

📅 28 April 2026, 15:00–16:30 (CET)
📍 Location: Campus Sterre, S5 Library room (third floor)

Target audience:

This session is for researchers who:

• write scripts or software as part of their work want their code to survive beyond their laptop
• use Git (or not yet…)
• are curious, but not necessarily experts If you already code but haven’t thought much about reproducibility, this is for you. This event is part of a new series organized by GhentCORR (Ghent University Community on Open and Reproducible Research). Learn more and connect with the community.

Agenda:

• Welcome and introduction
• Toon Verstraelen (Associate professor, Center for Molecular Modelling). Turning raw research results into a publication often becomes a scattered, non‑reproducible process across spreadsheets, scripts, and manuscript files. Toon Verstraelen will present how his team tackles this challenge with publication workflows that can reproduce the entire analysis starting from raw results all the way to the ZIP files with the manuscripts and datasets for upload to publishers and data repositories.
• Bart Mesuere (Assistant professor, Department of Mathematics, Computer Science and Statistics). Web-based research tools are convenient… until you need to reproduce last year’s results. Using Unipept (a research web application) as a case study, Bart Mesuere will discuss the reproducibility challenges that arise when both the software and its underlying reference data are regularly updated, and what practical guardrails help keep results traceable and repeatable over time.
• Sander Hendrickx (Lead Software Engineer, Biophysics Research Group). Developing a mature software package can be challenging when the team consists of PhD researchers rather than software developers. Sander Hendrickx will explain the hurdles of writing and maintaining a shared software framework with a team of researchers, and what they implemented to overcome these issues and streamline the development process.