Pseudotargeted Metabolomic Fingerprinting and Deep Learning for Identification and Visualization of Common Pathogens

Abstract

Matrix-assisted laser desorption/ionization time-of-flight mass (MALDI-TOF) spectrometry fingerprinting has reduced turnaround times, costs, and labor as conventional procedures in various laboratories. However, some species strains with high genetic correlation have not been directly distinguished using conventional standard procedures. Metabolomes can identify these strains by amplifying the minor differences because they are directly related to the phenotype. The pseudotargeted metabolomics method has the advantages of both non-targeted and targeted metabolomics. It can provide a new semi-quantitative fingerprinting with high coverage. We combined this pseudotargeted metabolomic fingerprinting with deep learning technology for the identification and visualization of the pathogen. A variational autoencoder framework was performed to identify and classify pathogenic bacteria and achieve their visualization, with prediction accuracy exceeding 99%. Therefore, this technology will be a powerful tool for rapidly and accurately identifying pathogens.

Semiquantitative Fingerprinting Based on Pseudotargeted Metabolomics and Deep Learning for the Identification of Listeria monocytogenes and Its Major Serotypes

“…monocytogenes, and the richer the metabolite information contained in the fingerprint, the better the identification accuracy. Then, the VAE-CNN model was used to identify pathogens . However, the differentiation for L.…”

“…Then, the VAE-CNN model was used to identify pathogens. 33 However, the differentiation for L. monocytogenes at the species and serotype levels were recognition failures due to the problems of overfitting and gradient explosion. Based on previous research, this study successfully established a new identification method for L. monocytogenes (up to the serotype level) using semiquantitative fingerprints based on LC−QQQ-MS combined with deep learning.…”

“…The combination of metabolite profiling and deep learning achieves the distinction of L. monocytogenes at the genus level. , In this study, we performed OPLS-DA analysis on the features of the whole metabolome, on the one hand, to select features that contribute specifically to the classification to form a characteristic fingerprint. Furthermore, the problems of overfitting and gradient explosion were solved by the ResNet model training.…”

Simultaneous Quantitation and Discovery (SQUAD) Analysis: Combining the Best of Targeted and Untargeted Mass Spectrometry-Based Metabolomics

“…In this study, the MRM ion pairs were also determined by searching relevant databases [ 48 ]. Another similar analysis, which is called pseudo-targeted metabolomics, was developed for the identification and visualization of common pathogens by analyzing a QC sample (mix of different bacterial strains) on a hybrid Quadrupole-Orbitrap mass spectrometer to select analytes for quantitation via an MRM analysis on TQ MS [ 49 ]. Deng et al also developed a pseudo-targeted approach to study metabolic changes in Asian plum ( Prunus salicina ) fruits in response to gummosis disease using a QTRAP for untargeted profiling followed by quantification of selected metabolites using MRM mode [ 50 ].…”

Simultaneous Quantitation and Discovery (SQUAD) Analysis: Combining the Best of Targeted and Untargeted Mass Spectrometry-Based Metabolomics

“…In this study, the MRM ion pairs were also determined by searching relevant databases [74]. Another similar analysis, which is called pseudo-targeted metabolomics, was developed for the identification and visualization of common pathogens by analyzing a QC sample (mix of different bacterial strains) on a hybrid Quadrupole-Orbitrap mass spectrometer to select analytes for quantitation via an MRM analysis on TQ MS [75]. Deng et al also developed a pseudo-targeted approach to study metabolic changes in Asian plum (Prunus salicina) fruits in response to gummosis disease using a QTRAP for untargeted profiling followed by quantification of selected metabolites using MRM mode [76].…”