FoodPro: A Web-Based Tool for Evaluating Covariance and Correlation NMR Spectra Associated with Food Processes

Area of Science:

• Food Science
• Analytical Chemistry
• Biotechnology

Background:

• Molecular mixture analysis during food processing offers insights into mechanisms for improved food quality.
• Existing tools lack comprehensive Nuclear Magnetic Resonance (NMR) spectral databases for diverse food processing applications.

Purpose of the Study:

• To develop a novel web-based tool, FoodPro, for accumulating and analyzing food NMR spectra.
• To correlate NMR spectral data with sensory (tasting) and physical (hardness) attributes of processed foods.

Main Methods:

• Development of a web-based platform integrating a database of 236 aqueous and 131 hydrophobic NMR spectra.
• Collection of 1753 tasting scores and 139 hardness measurements.
• Implementation of algorithms for querying NMR spectra, identifying similar spectra, and computing covariance and correlation spectra related to tasting and hardness.

Main Results:

• The FoodPro database contains NMR spectra, tasting scores, and hardness measurements, classified into fish and vegetable groups.
• The tool successfully queries user spectra, finds similar spectra, and estimates tasting and hardness.
• Analysis of fish spectra revealed specific covariance patterns related to tasting (e.g., lactate, glucose) and hardness (e.g., trimethylamine N-oxide).

Conclusions:

• FoodPro provides a valuable resource for molecular analysis of food processing.
• The tool demonstrates the potential of NMR spectroscopy to predict sensory and physical properties of food.
• Covariance analysis highlights specific molecules associated with taste and texture in processed foods.