Bacterial Wastewater-Based Epidemiology Using Surface-Enhanced Raman Spectroscopy and Machine Learning.

Area of Science:

• Environmental science
• Analytical chemistry
• Microbiology

Background:

• Wastewater-based epidemiology (WBE) is widely used for viral disease surveillance but less so for bacterial diseases.
• Distinguishing pathogenic from nonpathogenic bacteria using traditional PCR methods presents challenges.

Purpose of the Study:

• To develop a scalable, label-free method for detecting bacteria in wastewater using surface-enhanced Raman spectroscopy (SERS).
• To assess the potential of SERS for bacterial wastewater-based epidemiology.

Main Methods:

• Utilized plasmonic gold nanorods (AuNRs) to enhance Raman signals from bacteria in wastewater.
• Confirmed AuNR binding to bacterial surfaces via cryoelectron microscopy.
• Trained a machine learning model on spectral data from four clinically relevant bacterial species spiked in wastewater.

Main Results:

• Achieved over 87% accuracy in identifying bacterial species using the SERS-based machine learning model.
• Demonstrated an environmentally realistic limit of detection at 10^4 cells/mL.
• Validated SERS as a sensitive and specific method for bacterial detection in complex matrices.

Conclusions:

• SERS, enhanced by AuNRs, provides a promising label-free approach for bacterial detection in wastewater.
• This method represents a significant advancement toward establishing SERS platforms for bacterial WBE.
• The study overcomes limitations of PCR for bacterial identification in wastewater surveillance.