Abstract
Traditionally, specific bioreceptors such as antibodies have rapidly identified bacterial species in environmental water samples. However, this method has the disadvantages of requiring an additional process to conjugate or immobilize bioreceptors on the assay platform, which becomes unstable at room temperature. Here, we demonstrate a novel mix-and-match method to identify bacteria species by loading the bacterial samples with simple bacteria interacting components (not bioreceptors), such as lipopolysaccharides, peptidoglycan, and bovine serum albumin, and carboxylated particles, all separately on multiple channels. Neither covalent conjugation nor surface immobilization was necessary. Interactions between bacteria and the above bacteria interacting components resulted in varied surface tension and viscosity, leading to various flow velocities of capillary action through the paper fibers. The smartphone camera and a custom Python code recorded multiple channel flow velocity, each loaded with different bacteria interacting components. A multi-dimensional data set was obtained for a given bacterial species and concentration and used as a machine learning training model. A support vector machine was applied to classify the six bacterial species: Escherichia coli, Salmonella Typhimurium, Pseudomonas aeruginosa, Staphylococcus aureus, Enterococcus faecium, and Bacillus subtilis. Under optimized conditions, the training model predicts the bacterial species with an accuracy of > 85% of the six bacteria species.
Similar content being viewed by others
References
Sanganyado E, Gwenzi W. Antibiotic resistance in drinking water systems: occurrence, removal, and human health risks. Sci Total Environ. 2019;669:785–97. https://doi.org/10.1016/j.scitotenv.2019.03.162.
Sweeney RE, Budiman E, Yoon JY. Mie scatter spectra-based device for instant, contact-free, and specific diagnosis of bacterial skin infection. Sci Rep. 2017;7:4801. https://doi.org/10.1038/s41598-017-05061-1.
Angus SV, Cho S, Harshman DK, Song JY, Yoon JY. A Portable, shock-proof, surface-heated droplet PCR system for Escherichia coli detection. Biosens Bioelectron. 2015;74:360–8. https://doi.org/10.1016/j.bios.2015.06.026.
Day AS, Ulep TH, Budiman E, Dieckhaus L, Safavinia B, Hertenstein T, Yoon JY. Contamination-resistant, rapid emulsion-based isothermal nucleic acid amplification with Mie-scatter inspired light scatter analysis for bacterial identification. Sci Rep. 2021;11:19933. https://doi.org/10.1038/s41598-021-99200-4.
McCracken KE, Angus SV, Reynolds KA, Yoon JY. Multimodal imaging and lighting bias correction for improved µPAD-based water quality monitoring via smartphones. Sci Rep. 2016;6:27529. https://doi.org/10.1038/srep27529.
Klug KE, Reynolds KA, Yoon JY. A capillary flow dynamics-based sensing modality for direct environmental pathogen monitoring. Chem Eur J. 2018;24:6025–9. https://doi.org/10.1002/chem.201800085.
Kim S, Lee MH, Wiwasuku T, Day AS, Youngme S, Hwang DS, Yoon JY. Human sensor-inspired supervised machine learning of smartphone-based paper microfluidic analysis for bacterial species classification. Biosens Bioelectron. 2021;188:113335. https://doi.org/10.1016/j.bios.2021.113335.
Kwon HJ, Dean ZS, Angus SV, Yoon JY. Lab-on-a-chip for field Escherichia coli assays: long-term stability of reagents and automatic sampling system. JALA J Lab Automat. 2010;15:216–23. https://doi.org/10.1016/j.jala.2010.01.011.
Reitschel ET, Kirikae T, Shade FU, Mamat U, Schmidt G, Loppnow H, Ulmer AJ, Zähringer U, Seydel U, Padova FDI, Schreier M, Brade H. Bacterial endotoxin: molecular relationships of structure to activity and function. FASEB J. 1994;8:217–25. https://doi.org/10.1096/fasebj.8.2.8119492.
Atabek A, Liu Y, Pinzón-Arango PA, Camesano TA. Importance of LPS structure on protein interactions with Pseudomonas aeruginosa. Colloids Surf B Biointerf. 2008;67:115–21. https://doi.org/10.1016/j.colsurfb.2008.08.013.
Mena KD, Gerba CP. Risk assessment of Pseudomonas aeruginosa in water. Rev Environ Contam Toxicol. 2009;201:71–115. https://doi.org/10.1007/978-1-4419-0032-6_3.
Johnson JL, Ordal EJ. Deoxyribonucleic acid homology in bacterial taxonomy: effect of incubation temperature on reaction specificity. J Bacteriol. 1968;96:893–900. https://doi.org/10.1128/jb.95.3.893-900.1968.
Pandey PK, Kass PH, Soupir ML, Biswas S, Singh VP. Contamination of water resources by pathogenic bacteria. AMB Express. 2014;4:51. https://doi.org/10.1186/s13568-014-0051-x.
Kargalioglu Y, MacMillan BJ, Minear RA, Plewa MJ. Analysis of the cytotoxicity and mutagenicity of drinking water disinfection by-products in Salmonella Typhimurium. Teratog Carcinog Mutagen. 2002;22:113–28. https://doi.org/10.1002/tcm.10010.
Anastasi EM, Matthews B, Stratton HM, Katouli M. Pathogenic Escherichia coli found in sewage treatment plants and environmental waters. Appl Environ Microbiol. 2012;78:5536–41. https://doi.org/10.1128/AEM.00657-12.
Kim S, Romero-Lozano A, Hwang DS, Yoon JY. A guanidinium-rich polymer as a new universal bioreceptor for multiplex detection of bacteria from environmental samples. J Hazard Mater. 2021;413:125338. https://doi.org/10.1016/j.jhazmat.2021.125338.
Park TS, Li W, McCracken KE, Yoon JY. Smartphone quantifies Salmonella from paper microfluidics. Lab Chip. 2013;13:4832–40. https://doi.org/10.1039/c3lc50976a.
Acknowledgements
This work was supported by the University of Arizona’s Test All Test Smart program.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors declare no competing interests.
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
Below is the link to the electronic supplementary material.
Rights and permissions
About this article
Cite this article
Kim, S., Day, A.S. & Yoon, JY. Machine learning classification of bacterial species using mix-and-match reagents on paper microfluidic chips and smartphone-based capillary flow analysis. Anal Bioanal Chem 414, 3895–3904 (2022). https://doi.org/10.1007/s00216-022-04031-5
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00216-022-04031-5