How can we verify the origin of snow crabs, detect fraud in red seabream imports, or ensure the sustainability of wine and berry products or coffee—when we don’t know what to look for in advance? High-resolution mass spectrometry (HRMS) offers a powerful answer: non-target analysis (NTA). But the data are huge, complex, and often defy classical clustering or PCA.
This project explores Topological Data Analysis (TDA), a geometry-inspired approach to extract unique and robust chemical fingerprints from complex HRMS data. You’ll work with GC×GC/GC/LC–HRMS datasets from food and environmental samples, identify relatively unique mass spectral patterns, and visualize hidden relationships using Ball Mapper.
A chance to build interpretable models that link chemistry to traceability, origin, and sustainability—without needing a predefined target list.
Objectives of the Work (specific tasks will be agreed individually with the student)
• Understand preprocessing and vectorization of chromatogram data, GC×GC/GC/LC–HRMS.
• Implement Ball Mapper and statistically enhanced variants.
• Design uniqueness metrics and compare across product types (wine, crab, seabream, berries, coffee).
• Visualize chemical space with metadata overlays (e.g. origin, authenticity, sustainability).
• Compare TDA results with classical analysis (PCA, clustering, spectral similarity).
Benefits for the Student
• Learn how TDA bridges mathematics and real-world chemical data.
• Work with cutting-edge open metabolomics platforms.
• Lead in non-target analysis and sustainability science.
• Chace to submit a patent application and to multidisciplinary research.
Resources
- Yang, J., Shin, J., Kim, H., Sim, Y., & Yang, J. (2024). Discovery of candidate biomarkers to discriminate between Korean and Japanese red seabream (Pagrus major) using metabolomics. Food Chemistry, 431, 137129.
- Koljančić, N., Furdíková, K., de Araújo Gomes, A., & Špánik, I. (2024). Wine authentication: Current progress and state of the art. Trends in Food Science & Technology, 104598.
- Dłotko, P., Gurnari, D., & Sazdanovic, R. (2024). Mapper–type algorithms for complex data and relations. Journal of Computational and Graphical Statistics, 33(4), 1383-1396.
