Soha Farah

Composition of the jury:

• Mr. Alberto FIORE, Professor, Abertay University (UK), Reviewer
• Mr. Laurent DEBRAUWER, Senior Research Engineer, INRAE (Occitanie-Toulouse center), Reviewer
• Mrs. Nawel ACHIR, Professor, Institut Agro Montpellier, Examiner
• Mr. Samuel BERTRAND, Associate Professor, Nantes University, Examiner
• Mr. François FENAILLE,  Research Fellow, CEA (Université Paris Saclay), Examiner

Keywords: Thermal processing, untargeted approach, mass-spectrometry, feature-based molecular networking, chemical fingerprint

Abstract: Assessing food quality remains a major challenge due to the chemical complexity generated during processing. Thermal reactions such as the Maillard reaction, caramelisation, and lipid oxidation simultaneously produce compounds responsible for flavour and colour, but also for undesirable contaminants. Understanding how these reactions evolve and interact is therefore essential to advancing food quality design.

This work addresses this challenge by developing an untargeted analytical workflow combining LC HRMS and LC-HRMS/MS with Feature-based molecular networking (FBMN) to characterise and monitor chemical markers related to the different dimensions of food quality. The sponge cake was chosen as a model of formulated and thermally processed food, allowing controlled formulation and baking conditions.

After optimising extraction, acquisition parameters, and data treatment workflows, the approach was applied to sponge cakes baked under different conditions. FBMN highlighted compound families linked to Maillard and caramelisation processes and enabled the visualisation of reaction pathways. Formulation changes produced distinct molecular signatures, while a contamination scenario demonstrated the method’s ability to detect contaminants, although sensitivity at low levels remained limited.

Overall, the combination of LC-HRMS/MS with FBMN provided a comprehensive view of molecular transformations in thermally processed foods, offering a valuable framework for improving product safety, nutritional value and sensory attributes.