Machine learning-assisted universal PGM sensing platform: Recognition-detection separation via DNAzyme-nanozyme...

Area of Science:

• Analytical Chemistry
• Biotechnology
• Nanotechnology

Background:

• Personal glucose meters (PGMs) offer potential for point-of-care testing (POCT) beyond glucose monitoring.
• Matrix interference and the need for specific enzymes limit current PGM-based detection strategies.
• Developing universal and accurate detection methods is crucial for expanding PGM applications.

Purpose of the Study:

• To develop a highly sensitive and accurate PGM-based biosensor for ochratoxin A (OTA) detection.
• To create a versatile signal transduction method for PGM applications without glycosidases.
• To demonstrate the practical applicability of the biosensor for on-site mycotoxin detection.

Main Methods:

• Integration of a triplex molecular switch (THMS) with DNAzyme-mediated signal amplification.
• Utilized an Au@Pt nanozyme for signal transduction and Au@Pt nanoparticle release.
• Employed machine learning for biosensor performance analysis and four-parameter logistic regression for calibration.

Main Results:

• Achieved a linear detection range of 0-0.05 ng/mL for OTA with a detection limit of 0.008 ng/mL.
• The biosensor demonstrated a 10-fold lower detection limit compared to Au NPs-based methods.
• Obtained acceptable recovery rates (83.90%-116.66%) in corn and wine samples.

Conclusions:

• The developed PGM-based biosensor provides a portable and sensitive platform for on-site mycotoxin detection.
• The strategy of separating recognition and detection effectively eliminates matrix interference.
• This approach enhances PGM detection universality and accuracy, expanding POCT capabilities.