Rapid and Accurate Discrimination of Mycobacterium abscessus Subspecies Based on Matrix-Assisted Laser Desorption Ionization-Time of Flight Spectrum and Machine Learning Algorithms

Automated classification of group B Streptococcus into different clonal complexes using MALDI-TOF mass spectrometry

Objectives

To evaluate the performance of Matrix-Assisted Laser Desorption/Ionization Time-of Flight Mass Spectra (MALDI-TOF MS) for automated classification of GBS (Group B Streptococcus) into five major CCs (clonal complexes) during routine GBS identification.

Methods

MALDI-TOF MS of 167 GBS strains belonging to five major CCs (CC10, CC12, CC17, CC19, CC23) were grouped into a reference set (n = 67) and a validation set (n = 100) for the creation and evaluation with GBS CCs subtyping main spectrum (MSP) and MSP-M using MALDI BioTyper and ClinProTools. GBS CCs subtyping MSPs-M was generated by resetting the discriminative peaks of GBS CCs subtyping MSP according to the informative peaks from the optimal classification model of five major CCs and the contribution of each peak to the model created by ClinProTools.

Results

The PPV for the GBS CCs subtyping MSP-M was greater than the subtyping MSP for CC10 (99.21% vs. 93.65%), but similar for CC12 (79.55% vs. 81.06%), CC17 (93.55% vs. 94.09%), and CC19 (92.59% vs. 95.37%), and lower for CC23 (66.67% vs. 83.33%).

Conclusion

MALDI-TOF MS could be a promising tool for the automated categorization of GBS into 5 CCs by both CCs subtyping MSP and MSP-M, GBS CCs subtyping MSP-M is preferred for the accurate prediction of CCs with highly discriminative peaks.

Combining machine learning with high-content imaging to infer ciprofloxacin susceptibility in isolates of Salmonella Typhimurium

Antimicrobial resistance (AMR) is a growing public health crisis that requires innovative solutions. Current susceptibility testing approaches limit our ability to rapidly distinguish between antimicrobial-susceptible and -resistant organisms. Salmonella Typhimurium (S. Typhimurium) is an enteric pathogen responsible for severe gastrointestinal illness and invasive disease. Despite widespread resistance, ciprofloxacin remains a common treatment for Salmonella infections, particularly in lower-resource settings, where the drug is given empirically. Here, we exploit high-content imaging to generate deep phenotyping of S. Typhimurium isolates longitudinally exposed to increasing concentrations of ciprofloxacin. We apply machine learning algorithms to the imaging data and demonstrate that individual isolates display distinct growth and morphological characteristics that cluster by time point and susceptibility to ciprofloxacin, which occur independently of ciprofloxacin exposure. Using a further set of S. Typhimurium clinical isolates, we find that machine learning classifiers can accurately predict ciprofloxacin susceptibility without exposure to it or any prior knowledge of resistance phenotype. These results demonstrate the principle of using high-content imaging with machine learning algorithms to predict drug susceptibility of clinical bacterial isolates. This technique may be an important tool in understanding the morphological impact of antimicrobials on the bacterial cell to identify drugs with new modes of action.

Rapid discrimination of Bifidobacterium longum subspecies based on MALDI-TOF MS and machine learning

Although MALDI-TOF mass spectrometry (MS) is widely known as a rapid and cost-effective reference method for identifying microorganisms, its commercial databases face limitations in accurately distinguishing specific subspecies of Bifidobacterium. This study aimed to explore the potential of MALDI-TOF MS protein profiles, coupled with prediction methods, to differentiate between Bifidobacterium longum subsp. infantis (B. infantis) and Bifidobacterium longum subsp. longum (B. longum). The investigation involved the analysis of mass spectra of 59 B. longum strains and 41 B. infantis strains, leading to the identification of five distinct biomarker peaks, specifically at m/z 2,929, 4,408, 5,381, 5,394, and 8,817, using Recurrent Feature Elimination (RFE). To facilate classification between B. longum and B. infantis based on the mass spectra, machine learning models were developed, employing algorithms such as logistic regression (LR), random forest (RF), and support vector machine (SVM). The evaluation of the mass spectrometry data showed that the RF model exhibited the highest performace, boasting an impressive AUC of 0.984. This model outperformed other algorithms in terms of accuracy and sensitivity. Furthermore, when employing a voting mechanism on multi-mass spectrometry data for strain identificaton, the RF model achieved the highest accuracy of 96.67%. The outcomes of this research hold the significant potential for commercial applications, enabling the rapid and precise discrimination of B. longum and B. infantis using MALDI-TOF MS in conjunction with machine learning. Additionally, the approach proposed in this study carries substantial implications across various industries, such as probiotics and pharmaceuticals, where the precise differentiation of specific subspecies is essential for product development and quality control.