Protein Identification and Quantification Using Porous Silicon Arrays, Optical Measurements, and Machine Learning

Area of Science:

• Materials Science
• Biotechnology
• Sensor Technology

Background:

• Traditional biosensors often rely on specific capture agents, which can be expensive and limit versatility.
• Developing label-free detection methods is crucial for cost-effective and broadly applicable biosensing.

Purpose of the Study:

• To develop and demonstrate a versatile platform for analyte identification based on intrinsic physical and chemical properties.
• To showcase the ability of porous silicon (PSi) thin films to detect and discriminate proteins without capture agents.

Main Methods:

• Utilizing an array of porous silicon (PSi) thin films with varying pore sizes.
• Probing the reflectance of PSi array elements under different buffer pH conditions.
• Analyzing optical signals using machine learning algorithms for classification and quantification.

Main Results:

• Successfully classified, quantified, and discriminated three distinct proteins.
• Demonstrated reproducible protein identification and discrimination over a two-orders-of-magnitude concentration range.
• Achieved label-free detection by leveraging the physical and chemical properties of the analytes and the PSi sensor.

Conclusions:

• The PSi sensor array platform offers a versatile, low-cost, and robust approach to biomolecule detection.
• This method eliminates the need for expensive and limiting capture agents, paving the way for simpler biosensors.
• The study represents a significant advancement towards practical, agent-free biosensing applications.