Mass spectrometry and machine learning for the accurate diagnosis of benzylpenicillin and multidrug resistance of Staphylococcus aureus in bovine mastitis.

Staphylococcus aureus is a serious human and animal pathogen threat exhibiting extraordinary capacity for acquiring new antibiotic resistance traits in the pathogen population worldwide. The development of fast, affordable and effective diagnostic solutions capable of discriminating between antibiotic-resistant and susceptible S. aureus strains would be of huge benefit for effective disease detection and treatment. Here we develop a diagnostics solution that uses Matrix-Assisted Laser Desorption/Ionisation-Time of Flight Mass Spectrometry (MALDI-TOF) and machine learning, to identify signature profiles of antibiotic resistance to either multidrug or benzylpenicillin in S. aureus isolates. Using ten different supervised learning techniques, we have analysed a set of 82 S. aureus isolates collected from 67 cows diagnosed with bovine mastitis across 24 farms. For the multidrug phenotyping analysis, LDA, linear SVM, RBF SVM, logistic regression, naïve Bayes, MLP neural network and QDA had Cohen's kappa values over 85.00%. For the benzylpenicillin phenotyping analysis, RBF SVM, MLP neural network, naïve Bayes, logistic regression, linear SVM, QDA, LDA, and random forests had Cohen's kappa values over 85.00%. For the benzylpenicillin the diagnostic systems achieved up to (mean result ± standard deviation over 30 runs on the test set): accuracy = 97.54% ± 1.91%, sensitivity = 99.93% ± 0.25%, specificity = 95.04% ± 3.83%, and Cohen's kappa = 95.04% ± 3.83%. Moreover, the diagnostic platform complemented by a protein-protein network and 3D structural protein information framework allowed the identification of five molecular determinants underlying the susceptible and resistant profiles. Four proteins were able to classify multidrug-resistant and susceptible strains with 96.81% ± 0.43% accuracy. Five proteins, including the previous four, were able to classify benzylpenicillin resistant and susceptible strains with 97.54% ± 1.91% accuracy. Our approach may open up new avenues for the development of a fast, affordable and effective day-to-day diagnostic solution, which would offer new opportunities for targeting resistant bacteria.

Prediction of Streptococcus uberis clinical mastitis treatment success in dairy herds by means of mass spectrometry and machine-learning.

Streptococcus uberis is one of the leading pathogens causing mastitis worldwide. Identification of S. uberis strains that fail to respond to treatment with antibiotics is essential for better decision making and treatment selection. We demonstrate that the combination of supervised machine learning and matrix-assisted laser desorption ionization/time of flight (MALDI-TOF) mass spectrometry can discriminate strains of S. uberis causing clinical mastitis that are likely to be responsive or unresponsive to treatment. Diagnostics prediction systems trained on 90 individuals from 26 different farms achieved up to 86.2% and 71.5% in terms of accuracy and Cohen's kappa. The performance was further increased by adding metadata (parity, somatic cell count of previous lactation and count of positive mastitis cases) to encoded MALDI-TOF spectra, which increased accuracy and Cohen's kappa to 92.2% and 84.1% respectively. A computational framework integrating protein-protein networks and structural protein information to the machine learning results unveiled the molecular determinants underlying the responsive and unresponsive phenotypes.

Discrimination of contagious and environmental strains of Streptococcus uberis in dairy herds by means of mass spectrometry and machine-learning.

Streptococcus uberis is one of the most common pathogens of clinical mastitis in the dairy industry. Knowledge of pathogen transmission route is essential for the selection of the most suitable intervention. Here we show that spectral profiles acquired from clinical isolates using matrix-assisted laser desorption ionization/time of flight (MALDI-TOF) can be used to implement diagnostic classifiers based on machine learning for the successful discrimination of environmental and contagious S. uberis strains. Classifiers dedicated to individual farms achieved up to 97.81% accuracy at cross-validation when using a genetic algorithm, with Cohen's kappa coefficient of 0.94. This indicates the potential of the proposed methodology to successfully support screening at the herd level. A global classifier developed on merged data from 19 farms achieved 95.88% accuracy at cross-validation (kappa 0.93) and 70.67% accuracy at external validation (kappa 0.34), using data from another 10 farms left as holdout. This indicates that more work is needed to develop a screening solution successful at the population level. Significant MALDI-TOF spectral peaks were extracted from the trained classifiers. The peaks were found to correspond to bacteriocin and ribosomal proteins, suggesting that immunity, growth and competition over nutrients may be correlated to the different transmission routes.