Machine Learning-Assisted Pattern Recognition of Amyloid Beta Aggregates with Fluorescent Conjugated Polymers and...

Area of Science:

• Materials Science
• Biotechnology
• Analytical Chemistry

Background:

• Fluorescent conjugated polymers offer unique optical properties.
• Graphene oxide (GO) is a versatile nanomaterial with quenching capabilities.
• Developing sensitive and selective biosensors is crucial for disease diagnostics.

Purpose of the Study:

• To design and synthesize novel fluorescent poly(para-aryleneethynylene)s (P1-P5).
• To construct electrostatic complexes (C1-C5) with graphene oxide (GO) for biosensing applications.
• To evaluate the potential of these complexes for protein detection and Alzheimer's disease biomarker analysis.

Main Methods:

• Synthesis of five fluorescent poly(para-aryleneethynylene)s (P1-P5).
• Formation of electrostatic complexes (C1-C5) between positively charged polymers and negatively charged GO.
• Fluorescence quenching measurements to detect and differentiate analytes.
• Application of machine learning algorithms for data analysis and classification.

Main Results:

• The electrostatic complexes (C1-C5) exhibited fluorescence quenching upon interaction with GO.
• A sensor array comprising three complexes achieved 100% accuracy in distinguishing between 12 different proteins.
• The sensor array successfully identified different aggregation levels (monomers, oligomers, fibrils) of amyloid-beta 40 (Aβ40) and amyloid-beta 42 (Aβ42) using machine learning.

Conclusions:

• Fluorescent polymer-graphene oxide complexes are effective for sensitive protein detection.
• This sensor array technology shows great promise for the early diagnosis of Alzheimer's disease.
• The combination of advanced materials and machine learning offers a powerful strategy for complex biological sample analysis.