MSDeepAMR: antimicrobial resistance prediction based on deep neural networks and transfer learning.

Introduction: Antimicrobial resistance (AMR) is a global health problem that requires early and effective treatments to prevent the indiscriminate use of antimicrobial drugs and the outcome of infections. Mass Spectrometry (MS), and more particularly MALDI-TOF, have been widely adopted by routine clinical microbiology laboratories to identify bacterial species and detect AMR. The analysis of AMR with deep learning is still recent, and most models depend on filters and preprocessing techniques manually applied on spectra. Methods: This study propose a deep neural network, MSDeepAMR, to learn from raw mass spectra to predict AMR. MSDeepAMR model was implemented for Escherichia coli, Klebsiella pneumoniae, and Staphylococcus aureus under different antibiotic resistance profiles. Additionally, a transfer learning test was performed to study the benefits of adapting the previously trained models to external data. Results: MSDeepAMR models showed a good classification performance to detect antibiotic resistance. The AUROC of the model was above 0.83 in most cases studied, improving the results of previous investigations by over 10%. The adapted models improved the AUROC by up to 20% when compared to a model trained only with external data. Discussion: This study demonstrate the potential of the MSDeepAMR model to predict antibiotic resistance and their use on external MS data. This allow the extrapolation of the MSDeepAMR model to de used in different laboratories that need to study AMR and do not have the capacity for an extensive sample collection.

Value-added long-chain aliphatic compounds obtained through pyrolysis of phosphorylated chitin.

In this work, chitin, as a biobased polymer, is used as a precursor to obtain a phosphorylated derivatives. The influence of the different degree of phosphorylation in chitin on pyrolysis pattern was investigated. In order to understand the pyrolysis mechanism and the potential application of phosphorylated chitins, the samples were pyrolyzed at different temperatures and analyzed by FTIR, SEM, and Py-GC/MS analysis. Moreover, the thermal degradation and the evolved gases during chitin degradation and its derivatives were measured. The results showed that phosphorylation of chitin decreased the thermal stability of biopolymer and significantly changed the pattern of pyrolysis compared to neat chitin. The production of long-chain hydrocarbons was detected during pyrolysis of phosphorylated chitin, whereas this was not the case with raw chitin. Those two effects were more pronounced as the degree of phosphorylation increased. Chitin with the degree of phosphorylation (DS 1.35) exhibited the highest selectivity (91 %) towards production of long-chain hydrocarbons (C12-C17) at 500 °C. Moreover, the obtained results allowed to propose, for the first time, the mechanism of pyrolysis of phosphorylated chitin.