A novel short-term high-lactose culture approach combined with a matrix-assisted laser desorption ionization-time of flight mass spectrometry assay for differentiating Escherichia coli and Shigella species using artificial neural networks

A novel biochemistry approach combined with MALDI-TOF MS to discriminate Escherichia coli and Shigella species

Escherichia coli and Shigella spp. are closely related, making it crucial to accurately identify them for disease control and prevention. In this study, we utilized MALDI-TOF MS to identify characteristic peaks of decarboxylation products of lysine and ornithine to distinguish between E. coli and Shigella spp. Our findings indicate that the peak at m/z 103.12 ± 0.1 of the product cadaverine from lysine decarboxylase is unique to E. coli, while all Shigella species lack the m/z 103.12 ± 0.1 peak. However, S. sonnei and S. boydii serotype C13 exhibit a specific peak at m/z 89.10 ± 0.1, which is the product of putrescine from ornithine decarboxylase. We were able to correctly identify 97.06% (132 of 136) of E. coli and Shigella isolates and 100% (8 of 8) of S. sonnei isolates using this biochemical-based MALDI-TOF MS detection system. This technology is advantageous for its high-throughput, high quality, and ease of operation, and is of significant value for the diagnosis of E. coli and Shigella-related diseases.

Evaluation of the Zybio EXS3000 mass spectrometry in routine identification of Clinical isolates

The matrix-assisted laser desorption ionization-time of flight mass spectrometry (MALDI-TOF MS) has been widely applied in routine clinical microbiology laboratories as an efficient and reliable technique for diagnostic purpose. In this work, we evaluated the performance of the newly developed Zybio EXS3000 (Zybio Inc., China) in microbial identification and compared it with VITEK MS (bioMérieux, France). For this study, a total of 1340 isolates from various clinical specimens were collected. These isolates were analyzed simultaneously on both EXS3000 and VITEK MS. The inconsistent or unidentifiable data were further identified using the help of either 16S rRNA gene or ITS region sequencing. During the study, we observed that EXS3000 and VITEK MS provided positive confirmatory diagnostics for 95.0% and 96.5% of the isolates, respectively, which were consistent with the sequencing results. However, it is worth noting that the EXS3000 system needs to improve the identification performance of Candida albicans in the follow-up. There are no significant differences between the two devices in terms of microbial identification performance. The advantage of EXS3000 over VITEK MS is in its ability to perform in significantly lesser time period. In conclusion, the results of this investigation showed that EXS3000 can be used to identify microorganisms in clinical microbiology laboratories.

Non-lactose fermenting Escherichia coli: Following in the footsteps of lactose fermenting E. coli high-risk clones

Multi-resistant pathogenic strains of non-lactose fermenting Escherichia coli (NLF E. coli) are responsible for various intestinal and extraintestinal infections. Although several studies have characterised such strains using conventional methods, they have not been comprehensively studied at the genomic level. To address this gap, we used whole-genome sequencing (WGS) coupled with detailed microbiological and biochemical testing to investigate 17 NLF E. coli from a diagnostic centre (icddr,b) in Dhaka, Bangladesh. The prevalence of NLF E. coli was 10%, of which 47% (8/17) exhibited multi-drug resistant (MDR) phenotypes. All isolates (17/17) were confirmed as E. coli and could not ferment lactose sugar. WGS data analysis revealed international high-risk clonal lineages. The most prevalent sequence types (STs) were ST131 (23%), ST1193 (18%), ST12 (18%), ST501 (12%), ST167 (6%), ST73 (6%) and ST12 (6%). Phylogenetic analysis corroborated a striking clonal population amongst the studied NLF E. coli isolates. The predominant phylogroup detected was B2 (65%). The bla CTX-M-15 extended-spectrum beta-lactamase gene was present in 53% of isolates (9/17), whilst 64.7% (11/17) isolates were affiliated with pathogenic pathotypes. All extraintestinal pathogenic E. coli pathotypes demonstrated β-hemolysis. Our study underscores the presence of critical pathogens and MDR clones amongst non-lactose fermenting E. coli. We suggest that non-lactose fermenting E. coli be considered equally capable as lactose fermenting forms in causing intestinal and extraintestinal infections. Further, there is a need to undertake systematic, unbiased monitoring of predominant lineages amongst non-lactose fermenting E. coli that would help in better treatment and prevention strategies.

Bacterial species identification using MALDI-TOF mass spectrometry and machine learning techniques: A large-scale benchmarking study

Today machine learning methods are commonly deployed for bacterial species identification using MALDI-TOF mass spectrometry data. However, most of the studies reported in literature only consider very traditional machine learning methods on small datasets that contain a limited number of species. In this paper we present benchmarking results on an unprecedented scale for a wide range of machine learning methods, using datasets that contain almost 100,000 spectra and more than 1000 different species. The size and the diversity of the data allow to compare three important identification scenarios that are often not distinguished in literature, i.e., identification for novel biological replicates, novel strains and novel species that are not present in the training data. The results demonstrate that in all three scenarios acceptable identification rates are obtained, but the numbers are typically lower than those reported in studies with a more limited analysis. Using hierarchical classification methods, we also demonstrate that taxonomic information is in general not well preserved in MALDI-TOF mass spectrometry data. For the novel species scenario, we apply for the first time neural networks with Monte Carlo dropout, which have shown to be successful in other domains, such as computer vision, for the detection of novel species.

Investigation of MALDI-TOF Mass Spectrometry for Assessing the Molecular Diversity of Campylobacter jejuni and Comparison with MLST and cgMLST: A Luxembourg One-Health Study

There is a need for active molecular surveillance of human and veterinary Campylobacter infections. However, sequencing of all isolates is associated with high costs and a considerable workload. Thus, there is a need for a straightforward complementary tool to prioritize isolates to sequence. In this study, we proposed to investigate the ability of MALDI-TOF MS to pre-screen C. jejuni genetic diversity in comparison to MLST and cgMLST. A panel of 126 isolates, with 10 clonal complexes (CC), 21 sequence types (ST) and 42 different complex types (CT) determined by the SeqSphere+ cgMLST, were analysed by a MALDI Biotyper, resulting into one average spectra per isolate. Concordance and discriminating ability were evaluated based on protein profiles and different cut-offs. A random forest algorithm was trained to predict STs. With a 94% similarity cut-off, an AWC of 1.000, 0.933 and 0.851 was obtained for MLST_CC, MLST_ST and cgMLST profile, respectively. The random forest classifier showed a sensitivity and specificity up to 97.5% to predict four different STs. Protein profiles allowed to predict C. jejuni CCs, STs and CTs at 100%, 93% and 85%, respectively. Machine learning and MALDI-TOF MS could be a fast and inexpensive complementary tool to give an early signal of recurrent C. jejuni on a routine basis.

Historical, current, and emerging tools for identification and serotyping of Shigella

The Shigella genus includes serious foodborne disease etiologic agents, with 4 species and 54 serotypes. Identification at species and serotype levels is a crucial task in microbiological laboratories. Nevertheless, the genetic similarity between Shigella spp. and Escherichia coli challenges the correct identification and serotyping of Shigella spp., with subsequent negative repercussions on surveillance, epidemiological investigations, and selection of appropriate treatments. For this purpose, multiple techniques have been developed historically ranging from phenotype-based methods and single or multilocus molecular techniques to whole-genome sequencing (WGS). To facilitate the selection of the most relevant method, we herein provide a global overview of historical and emerging identification and serotyping techniques with a particular focus on the WGS-based approaches. This review highlights the excellent discriminatory power of WGS to more accurately elucidate the epidemiology of Shigella spp., disclose novel promising genomic targets for surveillance methods, and validate previous well-established methods.

Helix Matrix Transformation Combined With Convolutional Neural Network Algorithm for Matrix-Assisted Laser Desorption Ionization-Time of Flight Mass Spectrometry-Based Bacterial Identification

Matrix-assisted laser desorption ionization-time of flight mass spectrometry (MALDI-TOF MS) analysis is a rapid and reliable method for bacterial identification. Classification algorithms, as a critical part of the MALDI-TOF MS analysis approach, have been developed using both traditional algorithms and machine learning algorithms. In this study, a method that combined helix matrix transformation with a convolutional neural network (CNN) algorithm was presented for bacterial identification. A total of 14 bacterial species including 58 strains were selected to create an in-house MALDI-TOF MS spectrum dataset. The 1D array-type MALDI-TOF MS spectrum data were transformed through a helix matrix transformation into matrix-type data, which was fitted during the CNN training. Through the parameter optimization, the threshold for binarization was set as 16 and the final size of a matrix-type data was set as 25 × 25 to obtain a clean dataset with a small size. A CNN model with three convolutional layers was well trained using the dataset to predict bacterial species. The filter sizes for the three convolutional layers were 4, 8, and 16. The kernel size was three and the activation function was the rectified linear unit (ReLU). A back propagation neural network (BPNN) model was created without helix matrix transformation and a convolution layer to demonstrate whether the helix matrix transformation combined with CNN algorithm works better. The areas under the receiver operating characteristic (ROC) curve of the CNN and BPNN models were 0.98 and 0.87, respectively. The accuracies of the CNN and BPNN models were 97.78 ± 0.08 and 86.50 ± 0.01, respectively, with a significant statistical difference (p < 0.001). The results suggested that helix matrix transformation combined with the CNN algorithm enabled the feature extraction of the bacterial MALDI-TOF MS spectrum, which might be a proposed solution to identify bacterial species.

Recent evolutions of machine learning applications in clinical laboratory medicine

Machine learning (ML) is gaining increased interest in clinical laboratory medicine, mainly triggered by the decreased cost of generating and storing data using laboratory automation and computational power, and the widespread accessibility of open source tools. Nevertheless, only a handful of ML-based products are currently commercially available for routine clinical laboratory practice. In this review, we start with an introduction to ML by providing an overview of the ML landscape, its general workflow, and the most commonly used algorithms for clinical laboratory applications. Furthermore, we aim to illustrate recent evolutions (2018 to mid-2020) of the techniques used in the clinical laboratory setting and discuss the associated challenges and opportunities. In the field of clinical chemistry, the reviewed applications of ML algorithms include quality review of lab results, automated urine sediment analysis, disease or outcome prediction from routine laboratory parameters, and interpretation of complex biochemical data. In the hematology subdiscipline, we discuss the concepts of automated blood film reporting and malaria diagnosis. At last, we handle a broad range of clinical microbiology applications, such as the reduction of diagnostic workload by laboratory automation, the detection and identification of clinically relevant microorganisms, and the detection of antimicrobial resistance.

Prediction of malaria transmission drivers in Anopheles mosquitoes using artificial intelligence coupled to MALDI-TOF mass spectrometry

Vector control programmes are a strategic priority in the fight against malaria. However, vector control interventions require rigorous monitoring. Entomological tools for characterizing malaria transmission drivers are limited and are difficult to establish in the field. To predict Anopheles drivers of malaria transmission, such as mosquito age, blood feeding and Plasmodium infection, we evaluated artificial neural networks (ANNs) coupled to matrix-assisted laser desorption ionization-time of flight (MALDI-TOF) mass spectrometry (MS) and analysed the impact on the proteome of laboratory-reared Anopheles stephensi mosquitoes. ANNs were sensitive to Anopheles proteome changes and specifically recognized spectral patterns associated with mosquito age (0–10 days, 11–20 days and 21–28 days), blood feeding and P. berghei infection, with best prediction accuracies of 73%, 89% and 78%, respectively. This study illustrates that MALDI-TOF MS coupled to ANNs can be used to predict entomological drivers of malaria transmission, providing potential new tools for vector control. Future studies must assess the field validity of this new approach in wild-caught adult Anopheles. A similar approach could be envisaged for the identification of blood meal source and the detection of insecticide resistance in Anopheles and to other arthropods and pathogens.