Bacterial species identification from MALDI-TOF mass spectra through data analysis and machine learning

Area of Science:

• Microbiology
• Analytical Chemistry
• Bioinformatics

Background:

• Matrix-Assisted Laser Desorption/Ionization-Time of Flight Mass Spectrometry (MALDI-TOF MS) is a powerful tool for bacterial identification.
• Existing MALDI-TOF MS methodologies for bacterial characterization exhibit significant variability in sample preparation, matrix solutions, and data analysis.
• Standardization is crucial for reliable and reproducible bacterial identification using MALDI-TOF MS.

Purpose of the Study:

• To develop and validate a standardized MALDI-TOF MS protocol for the accurate identification of bacteria within the genera Leuconostoc, Fructobacillus, and Lactococcus.
• To assess the efficacy of the standardized protocol for species-level identification.
• To evaluate machine learning techniques as alternative tools for bacterial species identification.

Main Methods:

• A standardized protocol was developed using the CHCA matrix and a specific matrix solution (acetonitrile:water:trifluoroacetic acid).
• Bacterial strains were analyzed using both cell smear and cell extract methods.
• Data preprocessing, Principal Component Analysis (PCA), distance calculation, and multi-dimensional scaling were employed for analysis.
• Machine learning techniques, including support vector machines and random forests, were evaluated.

Main Results:

• The standardized MALDI-TOF MS protocol generated high-quality, species-specific mass spectra.
• Species identification within Leuconostoc and Fructobacillus achieved 84% accuracy using MALDI-TOF MS libraries.
• Identification within the Lactococcus genus reached 94% accuracy, considering species and subspecies.
• Machine learning models demonstrated high accuracies, ranging from 94% to 98%, for bacterial species identification.

Conclusions:

• The developed standardized MALDI-TOF MS protocol enables reliable and accurate species identification for Leuconostoc, Fructobacillus, and Lactococcus.
• MALDI-TOF MS, combined with advanced data analysis and machine learning, offers a robust approach for bacterial taxonomy.
• Further application of this standardized method can enhance microbial identification in various research and diagnostic settings.