A machine learning-enhanced biosensor for mercury detection based on an hydrophobin chimera

Area of Science:

• Environmental Science
• Biotechnology
• Analytical Chemistry

Background:

• Marine pollution from heavy metals, especially mercury, poses significant risks to human health.
• Conventional mercury detection methods are often costly, time-consuming, and require specialized equipment.
• There is a critical need for accessible and efficient tools for in situ water quality monitoring.

Purpose of the Study:

• To develop a novel, portable, and cost-effective biosensor for detecting mercury (II) in marine environments.
• To design a specific recognition element for mercury (II) that enables sensitive and selective detection.
• To integrate machine learning for simplified data interpretation and in situ analysis.

Main Methods:

• A hydrophobin-based chimera was engineered to bind mercury (II), causing a decrease in fluorescence.
• A fluorescence-based biosensor was constructed using the engineered chimera as the recognition element.
• Machine learning algorithms were employed to predict mercury concentrations without traditional readers.

Main Results:

• The biosensor achieved sensitive detection of mercury (II) in the nanomolar (nM) range.
• Detection limits were as low as 0.4 nM in tap water and 0.3 nM in seawater.
• The system demonstrated high specificity for Hg²⁺, even in the presence of other metal ions.
• Machine learning integration enabled mercury concentration prediction, facilitating in situ monitoring.

Conclusions:

• A novel, portable, and sensitive fluorescence-based biosensor for mercury (II) detection in seawater has been successfully developed.
• The biosensor, coupled with machine learning, offers a user-friendly and efficient platform for in situ environmental monitoring.
• This technology presents a significant advancement in addressing the challenge of mercury pollution in marine ecosystems.