Machine learning-based typing of Salmonella enterica O-serogroups by the Fourier-Transform Infrared (FTIR) Spectroscopy-based IR Biotyper system

A Comprehensive Methodology for Microbial Strain Typing Using Fourier-Transform Infrared Spectroscopy

Timely and accurate detection and characterization of microbial threats is crucial for effective infection and outbreak management. Additionally, in food production, rapid microbe identification is indispensable for maintaining quality control and hygiene standards. Current methods for typing microbial strains often rely on labor-intensive, time-consuming, and expensive DNA- and sera-serotyping techniques, limiting their applicability in rapid-response scenarios. In this context, the IR Biotyper®, utilizing Fourier-transform infrared (FTIR) spectroscopy, offers a novel approach, providing specific spectra for fast strain typing within 3 h. This methodology article serves as a comprehensive resource for researchers and technicians aiming to utilize FTIR spectroscopy for microbial strain typing. It encompasses detailed guidelines on sample preparation, data acquisition, and analysis techniques, ensuring the generation of reliable and reproducible results. We highlight the IR Biotyper®'s rapid and accurate discrimination capabilities, showcasing its potential for real-time pathogen monitoring and source-tracking to enhance public health and food safety. We propose its integration as an early screening method, followed by more detailed analysis with whole-genome sequencing, to optimize detection accuracy and response efficiency in microbial surveillance systems.

Bacterial discrimination by Fourier transform infrared spectroscopy, MALDI-mass spectrometry and whole-genome sequencing

Aim: Proof-of-concept study, highlighting the clinical diagnostic ability of FT-IR compared with MALDI-TOF MS, combined with WGS. Materials & methods: 104 pathogenic isolates of Neisseria meningitidis, Streptococcus pneumoniae, Streptococcus pyogenes and Staphylococcus aureus were analyzed. Results: Overall prediction accuracy was 99.6% in FT-IR and 95.8% in MALDI-TOF-MS. Analysis of N. meningitidis serogroups was superior in FT-IR compared with MALDI-TOF-MS. Phylogenetic relationship of S. pyogenes was similar by FT-IR and WGS, but not S. aureus or S. pneumoniae. Clinical severity was associated with the zinc ABC transporter and DNA repair genes in S. pneumoniae and cell wall proteins (biofilm formation, antibiotic and complement permeability) in S. aureus via WGS. Conclusion: FT-IR warrants further clinical evaluation as a promising diagnostic tool.

Strain typing and antimicrobial susceptibility of Salmonella enterica Albany isolates from duck farms in South Korea

Salmonella enterica is distributed worldwide and is a common cause of bacterial food poisoning in humans and a serious public health problem. Although duck meat consumption has recently increased in Korea, studies on the epidemiological relationship between S. enterica contamination in duck farms are scarce. Salmonella enterica serovar Albany isolates recovered from duck farms were analyzed using two typing methods - IR Biotyper® (IRBT) and multilocus variable-number tandem repeat analysis (MLVA). The clustering results were compared with the epidemiological survey findings and the antimicrobial resistance profiles. From April 2019 to October 2020, 20 individual feces per farm from 5-6-week-old ducks were collected repeatedly from 105 duck farms. Salmonella spp. isolated from duck feces were identified using PCR and multilocus sequence typing to investigate the prevalence and distribution of the Salmonella serovars. The prevalence of S. enterica was 19%, and S. enterica Albany was the predominantly recovered isolate. The S. enterica Albany isolates underwent antimicrobial susceptibility testing to determine the minimum inhibitory concentration. MLVA and IRBT methods established relatedness and diversity among the S. enterica Albany isolates. Multidrug-resistant S. enterica Albany was distributed in all the farms. Antimicrobial resistance profiles reflected the duck farm characteristics and isolates recovered from the same farm showed an identical profile. Isolates repeatedly recovered from the same farm also showed identical IRBT clusters and MLVA groups. These findings suggest that the isolates remained on the duck farm and re-infected new duck flocks. Thus, proper cleaning and disinfection is required before the farms are repopulated.

Development and validation of a quick, automated, and reproducible ATR FT-IR spectroscopy machine-learning model for Klebsiella pneumoniae typing

The reliability of Fourier-transform infrared (FT-IR) spectroscopy for Klebsiella pneumoniae typing and outbreak control has been previously assessed, but issues remain in standardization and reproducibility. We developed and validated a reproducible FT-IR with attenuated total reflectance (ATR) workflow for the identification of K. pneumoniae lineages. We used 293 isolates representing multidrug-resistant K. pneumoniae lineages causing outbreaks worldwide (2002-2021) to train a random forest classification (RF) model based on capsular (KL)-type discrimination. This model was validated with 280 contemporaneous isolates (2021-2022), using wzi sequencing and whole-genome sequencing as references. Repeatability and reproducibility were tested in different culture media and instruments throughout time. Our RF model allowed the classification of 33 capsular (KL)-types and up to 36 clinically relevant K. pneumoniae lineages based on the discrimination of specific KL- and O-type combinations. We obtained high rates of accuracy (89%), sensitivity (88%), and specificity (92%), including from cultures obtained directly from the clinical sample, allowing to obtain typing information the same day bacteria are identified. The workflow was reproducible in different instruments throughout time (>98% correct predictions). Direct colony application, spectral acquisition, and automated KL prediction through Clover MS Data analysis software allow a short time-to-result (5 min/isolate). We demonstrated that FT-IR ATR spectroscopy provides meaningful, reproducible, and accurate information at a very early stage (as soon as bacterial identification) to support infection control and public health surveillance. The high robustness together with automated and flexible workflows for data analysis provide opportunities to consolidate real-time applications at a global level. IMPORTANCE We created and validated an automated and simple workflow for the identification of clinically relevant Klebsiella pneumoniae lineages by FT-IR spectroscopy and machine-learning, a method that can be extremely useful to provide quick and reliable typing information to support real-time decisions of outbreak management and infection control. This method and workflow is of interest to support clinical microbiology diagnostics and to aid public health surveillance.

Application of Fourier Transform Infrared Spectroscopy to Discriminate Two Closely Related Bacterial Species: Bacillus anthracis and Bacillus cereus Sensu Stricto

Fourier transform infrared spectroscopy (FTIRS) is a diagnostic technique historically used in the microbiological field for the characterization of bacterial strains in relation to the specific composition of their lipid, protein, and polysaccharide components. For each bacterial strain, it is possible to obtain a unique absorption spectrum that represents the fingerprint obtained based on the components of the outer cell membrane. In this study, FTIRS was applied for the first time as an experimental diagnostic tool for the discrimination of two pathogenic species belonging to the Bacillus cereus group, Bacillus anthracis and Bacillus cereus sensu stricto; these are two closely related species that are not so easy to differentiate using classical microbiological methods, representing an innovative technology in the field of animal health.

Combining AI Tools with Non-Destructive Technologies for Crop-Based Food Safety: A Comprehensive Review

On a global scale, food safety and security aspects entail consideration throughout the farm-to-fork continuum, considering food's supply chain. Generally, the agrifood system is a multiplex network of interconnected features and processes, with a hard predictive rate, where maintaining the food's safety is an indispensable element and is part of the Sustainable Development Goals (SDGs). It has led the scientific community to develop advanced applied analytical methods, such as machine learning (ML) and deep learning (DL) techniques applied for assessing foodborne diseases. The main objective of this paper is to contribute to the development of the consensus version of ongoing research about the application of Artificial Intelligence (AI) tools in the domain of food-crop safety from an analytical point of view. Writing a comprehensive review for a more specific topic can also be challenging, especially when searching within the literature. To our knowledge, this review is the first to address this issue. This work consisted of conducting a unique and exhaustive study of the literature, using our TriScope Keywords-based Synthesis methodology. All available literature related to our topic was investigated according to our criteria of inclusion and exclusion. The final count of data papers was subject to deep reading and analysis to extract the necessary information to answer our research questions. Although many studies have been conducted, limited attention has been paid to outlining the applications of AI tools combined with analytical strategies for crop-based food safety specifically.

Successful autologous hematopoietic stem cell transplants using Salmonella positive products collected from asymptomatic donors

BACKGROUND

Bacterial contamination of hematopoietic stem cell (HSC) products is most commonly due to normal skin flora. Salmonella in HSC products is rare, and to our knowledge safe administration of an autologous HSC product containing Salmonella has not been reported.

STUDY DESIGN AND METHODS

We describe two patients undergoing autologous HSC transplant: peripheral blood HSC collection was performed by leukapheresis, and samples were cultured according to standard institutional protocol. Subsequent microorganism identification was performed using MALDI-TOF (Bruker Biotyper). Strain-relatedness was investigated by infrared spectroscopy using the IR Biotyper (Bruker).

RESULTS

The patients were asymptomatic throughout the collection process; however, HSC products collected on two consecutive days from each patient were positive for Salmonella. Isolates from both cultures were further characterized as Salmonella enterica serovar Dublin by the local public health department. Antibiotic susceptibility testing revealed different sensitivity patterns for the two strains. IR Biotyper demonstrated significant discriminatory power among the clinically significant Salmonella enterica subspecies, serogroups B, C1, and D. The patient strains were similar as both belonged to Group D Salmonella enterica serovar Dublin but were not identical. The Salmonella positive autologous HSC products were infused to both patients following administration of empiric antibiotic therapy. Both patients successfully engrafted and did well.

CONCLUSION

Salmonella is rarely seen in cellular therapy products and positivity may be the result of asymptomatic bacteremia at the time of collection. We present two instances of autologous HSC products containing Salmonella that were infused, along with prophylactic antimicrobial therapy without significant adverse clinical effects.

A multi-center validation study on the discrimination of Legionella pneumophila sg.1, Legionella pneumophila sg. 2-15 and Legionella non-pneumophila isolates from water by FT-IR spectroscopy

This study developed and validated a method, based on the coupling of Fourier-transform infrared spectroscopy (FT-IR) and machine learning, for the automated serotyping of Legionella pneumophila serogroup 1, Legionella pneumophila serogroups 2-15 as well as their successful discrimination from Legionella non-pneumophila. As Legionella presents significant intra- and inter-species heterogeneities, careful data validation strategies were applied to minimize late-stage performance variations of the method across a large microbial population. A total of 244 isolates were analyzed. In details, the method was validated with a multi-centric approach with isolates from Italian thermal and drinking water (n = 82) as well as with samples from German, Italian, French, and British collections (n = 162). Specifically, robustness of the method was verified over the time-span of 1 year with multiple operators and two different FT-IR instruments located in Italy and Germany. Moreover, different production procedures for the solid culture medium (in-house or commercial) and different culture conditions (with and without 2.5% CO₂) were tested. The method achieved an overall accuracy of 100, 98.5, and 93.9% on the Italian test set of Legionella, an independent batch of Legionella from multiple European culture collections, and an extra set of rare Legionella non-pneumophila, respectively.