Identification of β-lactam antibiotics using bioluminescent Escherichia coli and a support vector machine classifier algorithm

We propose the use of bioluminescent whole cell biosensor combined with a pattern classification algorithm to automatically detect and identify β-lactam antibiotic substances. Escherichia coli cells with a plasmid harboring luxCDABE genes under the β-lactam sensitive promoter element are used as sensors. We present experimental measurements of light production of bioluminescent bacteria subject to 11 antibiotic substances. The patterns of measured light production are classified using a support vector machine classifier. The accuracy and reliability of the classification suggests that this method can be used in the future to probe for new antibiotic substances.