A new scheme for strain typing of methicillin-resistant Staphylococcus aureus on the basis of matrix-assisted laser desorption ionization time-of-flight mass spectrometry by using machine learning approach

Abstract

Methicillin-resistant Staphylococcus aureus (MRSA), one of the most important clinical pathogens, conducts an increasing number of morbidity and mortality in the world. Rapid and accurate strain typing of bacteria would facilitate epidemiological investigation and infection control in near real time. Matrix-assisted laser desorption ionization-time of flight (MALDI-TOF) mass spectrometry is a rapid and cost-effective tool for presumptive strain typing. To develop robust method for strain typing based on MALDI-TOF spectrum, machine learning (ML) is a promising algorithm for the construction of predictive model. In this study, a strategy of building templates of specific types was used to facilitate generating predictive models of methicillin-resistant Staphylococcus aureus (MRSA) strain typing through various ML methods. The strain types of the isolates were determined through multilocus sequence typing (MLST). The area under the receiver operating characteristic curve (AUC) and the predictive accuracy of the models were compared. ST5, ST59, and ST239 were the major MLST types, and ST45 was the minor type. For binary classification, the AUC values of various ML methods ranged from 0.76 to 0.99 for ST5, ST59, and ST239 types. In multiclass classification, the predictive accuracy of all generated models was more than 0.83. This study has demonstrated that ML methods can serve as a cost-effective and promising tool that provides preliminary strain typing information about major MRSA lineages on the basis of MALDI-TOF spectra.

Rapidly predicting vancomycin resistance ofEnterococcus faeciumthrough MALDI-TOF MS spectrum obtained in real-world clinical microbiology laboratory

“…14–16 On the basis of the massive data produced by MALDI-TOF MS, moreover, some studies have demonstrated that subspecies typing could be predicted from a specific pattern of MS spectra only. 17,19 Furthermore, other studies have shown a good correlation between AST and specific patterns of MS spectra. 18,23–25,48 However, some issues have limited the generalization of these results.…”

“…A log score (generated through Biotyper 3.1) larger than 2 was used for confirming E. faecium . 17–19 We tested vancomycin susceptibility of E. faecium by using the paper disc method. The details of E. faecium identification and AST are given in the eMethods in the Supplement.…”

“…These ML algorithms have demonstrated their successful applications (either classification or prediction) in clinical practice. 17–19,27,28,35,36 The details of these ML algorithms and model training processes are attached in the eMethods in the Supplement.…”

“…14–16 In addition to species identification, MALDI-TOF MS has been promising in other applications, such as strain typing or AST. 17–19 MALDI-TOF MS can generate massive data comprising hundreds of peak signals on the spectra. 17,20 The complex data of MALDI-TOF spectra are overwhelming to even an experienced medical staff.…”

“…17,20 The complex data of MALDI-TOF spectra are overwhelming to even an experienced medical staff. 19 Studies have attempted to identify the characteristic peak through visual inspection. 21,22 The results of the studies have been discordant, which has limited the clinical utility.…”