A simple array integrating machine learning for identification of flavonoids in red wines

Area of Science:

• Analytical Chemistry
• Biochemistry
• Materials Science

Background:

• Bioactive flavonoids in red wine offer cardiovascular benefits but are difficult to identify.
• Existing methods for flavonoid detection are often complex and time-consuming.

Purpose of the Study:

• To develop a simple yet effective sensor array for the detection and discrimination of flavonoids.
• To assess the array's capability in analyzing complex mixtures like red wine.

Main Methods:

• Fabrication of a sensor array using phenylboronic acid modified perylene diimide derivatives (PDIs).
• Utilizing multiple non-specific interactions for flavonoid binding and fluorescence modulation.
• Employing machine learning algorithms to analyze fluorescence intensity patterns for identification.

Main Results:

• The PDI sensor array accurately discriminated 14 diverse flavonoids with 100% accuracy.
• The array successfully performed parallel detection of 8 different red wine types and origins.
• Fluorescence modulation patterns generated unique fingerprints for each flavonoid.

Conclusions:

• The developed sensor array offers a highly accurate and efficient method for flavonoid detection.
• This technology shows significant potential for analyzing complex food matrices, including red wine.
• The array's ability to differentiate wine types highlights its application in food authentication and quality control.