Rapid classification of heavy metal-exposed freshwater bacteria by infrared spectroscopy coupled with chemometrics using supervised method

Gurbanov R., Gözen A. G., Severcan F.

SPECTROCHIMICA ACTA PART A-MOLECULAR AND BIOMOLECULAR SPECTROSCOPY, vol.189, pp.282-290, 2018 (SCI-Expanded) identifier identifier identifier

  • Publication Type: Article / Article
  • Volume: 189
  • Publication Date: 2018
  • Doi Number: 10.1016/j.saa.2017.08.038
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Page Numbers: pp.282-290
  • Keywords: Bacteria classification, Chemometric methods, Soft independent modeling of class analogy (SIMCA), IR spectroscopy, FT-IR SPECTROSCOPY, STAPHYLOCOCCUS-AUREUS, ESCHERICHIA-COLI, IDENTIFICATION, DISCRIMINATION, BIOREMEDIATION, STRAINS, FOOD, POPULATIONS, PATHOGENS
  • Middle East Technical University Affiliated: Yes


Rapid, cost-effective, sensitive and accurate methodologies to classify bacteria are still in the process of development. The major drawbacks of standard microbiological, molecular and immunological techniques call for the possible usage of infrared (IR) spectroscopy based supervised chemometric techniques. Previous applications of IR based chemometric methods have demonstrated outstanding findings in the classification of bacteria. Therefore, we have exploited an IR spectroscopy based chemometrics using supervised method namely Soft Independent Modeling of Class Analogy (SIMCA) technique for the first time to classify heavy metal-exposed bacteria to be used in the selection of suitable bacteria to evaluate their potential for environmental cleanup applications. Herein, we present the powerful differentiation and classification of laboratory strains (Escherichia coli and Staphylococcus aureus) and environmental isolates (Gordonia sp. and Microbacterium oxydans) of bacteria exposed to growth inhibitory concentrations of silver (Ag), cadmium (Cd) and lead (Pb). Our results demonstrated that SIMCA was able to differentiate all heavy metal-exposed and control groups from each other with 95% confidence level. Correct identification of randomly chosen test samples in their corresponding groups and high model distances between the classes were also achieved. We report, for the first time, the success of IR spectroscopy coupled with supervised chemometric technique SIMCA in classification of different bacteria under a given treatment. (C) 2017 Elsevier B.V. All rights reserved.