Ratiometric, 3D Fluorescence Spectrum with Abundant Information for Tetracyclines Discrimination via Dual...

Area of Science:

• Analytical Chemistry
• Biotechnology
• Machine Learning

Background:

• Tetracyclines are crucial antibiotics for bacterial infections.
• Distinguishing between similar tetracycline structures using biosensors is challenging due to subtle chemical differences.
• Existing single-probe methods often suffer from information overlap in spectral analysis.

Purpose of the Study:

• To develop a novel strategy for accurate discrimination of tetracyclines.
• To leverage aptamers and 3D fluorescence spectroscopy for enhanced spectral fingerprinting.
• To apply deep learning for qualitative and quantitative analysis of tetracyclines.

Main Methods:

• Utilized aptamers for dual biomolecule recognition to generate distinct ratiometric, 3D fluorescence spectra for each tetracycline.
• Employed an artificial neural network model to process the complex spectral fingerprint information.
• Compared the performance of the dual biomolecule recognition strategy against conventional single-probe methods.

Main Results:

• Successfully generated distinct, ratiometric, 3D fluorescence spectra for individual tetracyclines.
• Achieved accurate qualitative and quantitative analysis of tetracyclines using the developed deep learning model.
• Demonstrated significantly higher accuracy compared to single-probe methods.

Conclusions:

• The dual biomolecule recognition strategy combined with ratiometric 3D fluorescence spectroscopy provides rich fingerprint information for deep learning.
• This approach offers a new, highly accurate method for discriminating analytes, particularly challenging antibiotic compounds like tetracyclines.
• The findings pave the way for advanced 3D fluorescence-based analytical techniques.