UNLOCKING THE FUTURE OF COMPREHENSIVE TWO-DIMENSIONAL CHROMATOGRAPHY IN FOOD-OMICS BY ARTIFICIAL INTELLIGENCE ALGORITHMS

Since its introduction, comprehensive two-dimensional gas chromatography (GC×GC), has unrevealed its potentials in many fields helping scientists to better understand the Nature’s complexity, facilitating highly-informative screenings, supporting markers discovery in omics applications and offering many opportunities to implement system biology-like strategies for investigation, the so-called integrationist approach [1]. In food “omics” the analytical platform design and configuration plays a key role to achieve the suitable information capacity, resolution and sensitivity to answer the many questions posed by application needs. The contribution deals with the challenging task of designing a multidimensional platform for food metabolomics [2] implemented by an effective data processing workflow. A strategy capable to answer many questions about product qualities (e.g., sensory quality, freshness, authenticity, presence of sensory defects etc.) with a single measure realized by combining many analytical dimensions (e.g., sample preparation, separation, multiple detection, olfactometry, etc.). Within this context, the key-role of Artificial Intelligence (AI) algorithms for computer vision (i.e., “…a field of AI that enables computers and systems to derive meaningful information from digital images…” [3]) and smelling (e.g., AI smelling machine [4]) is discussed and proof-of-evidence on the feasibility and effectiveness of such “comprehensive” approaches presented through the authors research experience on high-quality extra-virgin olive oil.