Evolved microbial diversity enables combinatoric biosensing in complex environments.

Area of Science:

• Environmental Science
• Biotechnology
• Machine Learning

Background:

• Whole-cell biosensors (WCBs) are effective for detecting single environmental contaminants in labs.
• WCBs need to be adapted for complex environments to monitor multiple targets simultaneously.

Purpose of the Study:

• To develop a generalizable framework for quantifying multiple analytes using WCBs.
• To leverage supervised machine learning and sensor cross-specificity for multi-analyte detection.

Main Methods:

• Engineered six single-target heavy metal sensors in E. coli.
• Evolved E. coli strains for improved seawater growth, creating 30 sensor variants.
• Utilized microfluidics to characterize variant responses and applied machine learning to a sensor consortium.

Main Results:

• Characterized significant dynamic response diversity across 30 WCB variants.
• Developed a consortium approach that combinatorically quantifies multiple analytes.
• Achieved superior performance over single-target biosensors in over 90% of test samples.

Conclusions:

• The proposed framework enables WCBs to quantify multiple targets in complex environments.
• This approach facilitates the translation of WCB technology beyond laboratory conditions.
• The study establishes a scalable method for advanced environmental monitoring.