Automated-Screening Oriented Electric Sensing of Vitamin B1 Using a Machine Learning Aided Solid-State Nanopore

Area of Science:

• Nanotechnology and biosensing
• Machine learning in diagnostics
• Analytical chemistry

Background:

• Accurate micronutrient detection is crucial for clinical and home diagnostics.
• Current methods like HPLC and LC-MS are costly and time-consuming.
• There is a need for rapid, cost-effective diagnostic tools.

Purpose of the Study:

• To develop an automated electric sensing strategy for single-molecule detection of vitamin B1 and its derivatives.
• To explore the relationship between vitamin B1 dynamics and nanopore signatures.
• To validate the use of machine learning in this diagnostic approach.

Main Methods:

• Utilizing solid-state nanopore technology for electric sensing.
• Employing machine learning algorithms to analyze nanopore signatures.
• Investigating vitamin B1 dynamics and their correlation with electrical signals.
• Applying Shapley additive explanations to interpret machine learning decisions.

Main Results:

• Achieved accurate identification of vitamin B1 and its phosphorylated derivatives.
• Demonstrated a strong correlation between vitamin B1 dynamics and observed nanopore signatures.
• Successfully interpreted the machine learning decision-making process.

Conclusions:

• The proposed automated electric sensing strategy is a viable next-generation approach for micronutrient detection.
• Merging nanopore signatures with machine learning offers a powerful tool for diagnostics.
• This method presents a faster and potentially more affordable alternative to conventional techniques.