Identification of antibiotic resistance and virulence-encoding factors in Klebsiella pneumoniae by Raman spectroscopy and deep learning

Identification of hypermucoviscous Klebsiella pneumoniae K1, K2, K54 and K57 capsular serotypes by Raman spectroscopy

Antimicrobial resistance poses a significant challenge in modern medicine, affecting public health. Klebsiella pneumoniae infections compound this issue due to their broad range of infections and the emergence of multiple antibiotic resistance mechanisms. Efficient detection of its capsular serotypes is crucial for immediate patient treatment, epidemiological tracking and outbreak containment. Current methods have limitations that can delay interventions and increase the risk of morbidity and mortality. Raman spectroscopy is a promising alternative to identify capsular serotypes in hypermucoviscous K. pneumoniae isolates. It provides rapid and in situ measurements with minimal sample preparation. Moreover, its combination with machine learning tools demonstrates high accuracy and reproducibility. This study analyzed the viability of combining Raman spectroscopy with one-dimensional convolutional neural networks (1-D CNN) to classify four capsular serotypes of hypermucoviscous K. pneumoniae: K1, K2, K54 and K57. Our approach involved identifying the most relevant Raman features for classification to prevent overfitting in the training models. Simplifying the dataset to essential information maintains accuracy and reduces computational costs and training time. Capsular serotypes were classified with 96 % accuracy using less than 30 Raman features out of 2400 contained in each spectrum. To validate our methodology, we expanded the dataset to include both hypermucoviscous and non-mucoid isolates and distinguished between them. This resulted in an accuracy rate of 94 %. The results obtained have significant potential for practical healthcare applications, especially for enabling the prompt prescription of the appropriate antibiotic treatment against infections.

Implications of Artificial Intelligence in Addressing Antimicrobial Resistance: Innovations, Global Challenges, and Healthcare's Future

Antibiotic resistance poses a significant threat to global public health due to complex interactions between bacterial genetic factors and external influences such as antibiotic misuse. Artificial intelligence (AI) offers innovative strategies to address this crisis. For example, AI can analyze genomic data to detect resistance markers early on, enabling early interventions. In addition, AI-powered decision support systems can optimize antibiotic use by recommending the most effective treatments based on patient data and local resistance patterns. AI can accelerate drug discovery by predicting the efficacy of new compounds and identifying potential antibacterial agents. Although progress has been made, challenges persist, including data quality, model interpretability, and real-world implementation. A multidisciplinary approach that integrates AI with other emerging technologies, such as synthetic biology and nanomedicine, could pave the way for effective prevention and mitigation of antimicrobial resistance, preserving the efficacy of antibiotics for future generations.

Genome analyses of colistin-resistant high-risk blaNDM-5 producing Klebsiella pneumoniae ST147 and Pseudomonas aeruginosa ST235 and ST357 in clinical settings

BACKGROUND

Colistin is a last-resort antibiotic used in extreme cases of multi-drug resistant (MDR) Gram-negative bacterial infections. Colistin resistance has increased in recent years and often goes undetected due to the inefficiency of predominantly used standard antibiotic susceptibility tests (AST). To address this challenge, we aimed to detect the prevalence of colistin resistance strains through both Vitek®2 and broth micro-dilution. We investigated 1748 blood, tracheal aspirate, and pleural fluid samples from the Intensive Care Unit (ICU), Neonatal Intensive Care Unit (NICU), and Tuberculosis and Respiratory Disease centre (TBRD) in an India hospital. Whole-genome sequencing (WGS) of extremely drug-resitant (XDR) and pan-drug resistant (PDR) strains revealed the resistance mechanisms through the Resistance Gene Identifier (RGI.v6.0.0) and Snippy.v4.6.0. Abricate.v1.0.1, PlasmidFinder.v2.1, MobileElementFinder.v1.0.3 etc. detected virulence factors, and mobile genetic elements associated to uncover the pathogenecity and the role of horizontal gene transfer (HGT).

RESULTS

This study reveals compelling insights into colistin resistance among global high-risk clinical isolates: Klebsiella pneumoniae ST147 (16/20), Pseudomonas aeruginosa ST235 (3/20), and ST357 (1/20). Vitek®2 found 6 colistin-resistant strains (minimum inhibitory concentrations, MIC = 4 μg/mL), while broth microdilution identified 48 (MIC = 32-128 μg/mL), adhering to CLSI guidelines. Despite the absence of mobile colistin resistance (mcr) genes, mechanisms underlying colistin resistance included mgrB deletion, phosphoethanolamine transferases arnT, eptB, ompA, and mutations in pmrB (T246A, R256G) and eptA (V50L, A135P, I138V, C27F) in K. pneumoniae. P. aeruginosa harbored phosphoethanolamine transferases basS/pmrb, basR, arnA, cprR, cprS, alongside pmrB (G362S), and parS (H398R) mutations. Both strains carried diverse clinically relevant antimicrobial resistance genes (ARGs), including plasmid-mediated blaNDM-5 (K. pneumoniae ST147) and chromosomally mediated blaNDM-1 (P. aeruginosa ST357).

CONCLUSION

The global surge in MDR, XDR and PDR bacteria necessitates last-resort antibiotics such as colistin. However, escalating resistance, particularly to colistin, presents a critical challenge. Inefficient colistin resistance detection methods, including Vitek2, alongside limited surveillance resources, accentuate the need for improved strategies. Whole-genome sequencing revealed alarming colistin resistance among K. pneumoniae and P. aeruginosa in an Indian hospital. The identification of XDR and PDR strains underscores urgency for enhanced surveillance and infection control. SNP analysis elucidated resistance mechanisms, highlighting the complexity of combatting resistance.

Tackling the Antimicrobial Resistance "Pandemic" with Machine Learning Tools: A Summary of Available Evidence

Antimicrobial resistance is recognised as one of the top threats healthcare is bound to face in the future. There have been various attempts to preserve the efficacy of existing antimicrobials, develop new and efficient antimicrobials, manage infections with multi-drug resistant strains, and improve patient outcomes, resulting in a growing mass of routinely available data, including electronic health records and microbiological information that can be employed to develop individualised antimicrobial stewardship. Machine learning methods have been developed to predict antimicrobial resistance from whole-genome sequencing data, forecast medication susceptibility, recognise epidemic patterns for surveillance purposes, or propose new antibacterial treatments and accelerate scientific discovery. Unfortunately, there is an evident gap between the number of machine learning applications in science and the effective implementation of these systems. This narrative review highlights some of the outstanding opportunities that machine learning offers when applied in research related to antimicrobial resistance. In the future, machine learning tools may prove to be superbugs' kryptonite. This review aims to provide an overview of available publications to aid researchers that are looking to expand their work with new approaches and to acquaint them with the current application of machine learning techniques in this field.

Siamese Networks for Clinically Relevant Bacteria Classification Based on Raman Spectroscopy

Identifying bacterial strains is essential in microbiology for various practical applications, such as disease diagnosis and quality monitoring of food and water. Classical machine learning algorithms have been utilized to identify bacteria based on their Raman spectra. However, convolutional neural networks (CNNs) offer higher classification accuracy, but they require extensive training sets and retraining of previous untrained class targets can be costly and time-consuming. Siamese networks have emerged as a promising solution. They are composed of two CNNs with the same structure and a final network that acts as a distance metric, converting the classification problem into a similarity problem. Classical machine learning approaches, shallow and deep CNNs, and two Siamese network variants were tailored and tested on Raman spectral datasets of bacteria. The methods were evaluated based on mean sensitivity, training time, prediction time, and the number of parameters. In this comparison, Siamese-model2 achieved the highest mean sensitivity of 83.61 ± 4.73 and demonstrated remarkable performance in handling unbalanced and limited data scenarios, achieving a prediction accuracy of 73%. Therefore, the choice of model depends on the specific trade-off between accuracy, (prediction/training) time, and resources for the particular application. Classical machine learning models and shallow CNN models may be more suitable if time and computational resources are a concern. Siamese networks are a good choice for small datasets and CNN for extensive data.

Metagenomics next-generation sequencing (mNGS) reveals emerging infection induced by Klebsiella pneumoniaeniae

OBJECTIVES

The increasing emergence of hypervirulent Klebsiella pneumoniae (hv-Kp) and carbapenem-resistant K. pneumoniae (CR-Kp) is a serious and substantial public health problem. The use of the last resort antimicrobials, tigecycline and polymyxin to combat infections is complicated by the expanding repertoire of newly-identified CR-hvKp. The transmission and co-occurrence of the corresponding antimicrobial resistance and virulence determinants are largely unknown. The aim of this study was to investigate the dissemination and dynamics of CR-Kp and its antibiotic resistance in a hospitalised patient.

METHODS

Metagenomic next-generation sequencing (mNGS) was conducted for different specimens collected from an elderly male hospitalised patient. CR-Kp strains were examined using antibiotic susceptibility and string testing. Antimicrobial and virulence genes were annotated using whole-genome sequencing (WGS).

RESULTS

A clinical case of a patient infected with a variety of CR-Kp isolates was reported. The co-occurrence of KPC-2 and NDM-1 in the patient was revealed. The CR-Kp isolates, such as BALF2, and Sputum T1 and T3, were classified into ST11 and ST147, respectively. The genetic signature (iuc operon) of hypervirulence was identified in strain T1, although string testing indicated its intermediate virulence.

CONCLUSIONS

In this study, multiple infections of CR-Kp isolates were revealed by mNGS, and their dissemination was attributed to plasmid variations, mgrB inactivation and integrative conjugative elements (ICEs). Furthermore, the finding indicated one likely convergence to form CR-hvKp, different from acquisition of carbapenem-resistance determinants in hvKp. A combination of mNGS and WGS is beneficial for clinical diagnosis and anti-infection therapy, and facilitates a better understanding of genetic variants conferring antimicrobial and virulence properties.

Machine learning model of imipenem-resistant Klebsiella pneumoniae based on MALDI-TOF-MS platform: An observational study

Background and Aim

Machine learning is an important branch and supporting technology of artificial intelligence, we established four machine learning model for the drug sensitivity of Klebsiella pneumoniae to imipenem based on matrix-assisted laser desorption ionization time-of-flight mass spectrometry (MALDI-TOF-MS) and compared their diagnostic effect.

Methods

The data of MALDI-TOF-MS and imipenem sensitivity of 174 cases of K. pneumoniae isolated from clinical specimens in the laboratory of microbiology department of Tianjin Haihe Hospital from 2019 January to 2020 December were collected. The mass spectrometry and imipenem sensitivity of 70 cases of imipenem-sensitive and 70 resistant cases were randomly selected to establish the training set model, 17 cases of sensitive and 17 cases of resistant cases were randomly selected to establish the test set model. Mass spectral peak data were subjected to orthogonal partial least squares discriminant analysis (OPLS-DA), the training set data model was established by machine learning least absolute shrinkage and selection operator (LASSO) algorithm, logistic regression (LR) algorithm, support vector machines (SVM) algorithm, neural network (NN) algorithm, the area under the curve (AUC) and confusion matrix of training set and test set model were calculated and selected by Grid search and 3-fold Cross-validation respectively, the accuracy of the prediction model was verified by test set confusion matrix.

Results

The R²Y and Q² of OPLS-DA were 0.546 and 0.0178. The AUC of the best training set and test set models were 0.9726 and 0.9100, 1.0000 and 0.8581, 0.8462 and 0.6263, 1.0000 and 0.7180 evaluated by LASSO, LR, SVM and NN model respectively. The accuracy of the LASSO, LR, SVM and NN model were 87%, 79%, 62%, and 68% in test set, respectively.

Conclusion

The LASSO prediction model of K. pneumoniae sensitivity to imipenem established in this study has a high accuracy rate and has potential clinical decision support ability.

Rapid Prediction of Multidrug-Resistant Klebsiella pneumoniae through Deep Learning Analysis of SERS Spectra

Klebsiella pneumoniae is listed by the WHO as a priority pathogen of extreme importance that can cause serious consequences in clinical settings. Due to its increasing multidrug resistance all over the world, K. pneumoniae has the potential to cause extremely difficult-to-treat infections. Therefore, rapid and accurate identification of multidrug-resistant K. pneumoniae in clinical diagnosis is important for its prevention and infection control. However, the limitations of conventional and molecular methods significantly hindered the timely diagnosis of the pathogen. As a label-free, noninvasive, and low-cost method, surface-enhanced Raman scattering (SERS) spectroscopy has been extensively studied for its application potentials in the diagnosis of microbial pathogens. In this study, we isolated and cultured 121 K. pneumoniae strains from clinical samples with different drug resistance profiles, which included polymyxin-resistant K. pneumoniae (PRKP; n = 21), carbapenem-resistant K. pneumoniae, (CRKP; n = 50), and carbapenem-sensitive K. pneumoniae (CSKP; n = 50). For each strain, a total of 64 SERS spectra were generated for the enhancement of data reproducibility, which were then computationally analyzed via the convolutional neural network (CNN). According to the results, the deep learning model CNN plus attention mechanism could achieve a prediction accuracy as high as 99.46%, with robustness score of 5-fold cross-validation at 98.87%. Taken together, our results confirmed the accuracy and robustness of SERS spectroscopy in the prediction of drug resistance of K. pneumoniae strains with the assistance of deep learning algorithms, which successfully discriminated and predicted PRKP, CRKP, and CSKP strains. IMPORTANCE This study focuses on the simultaneous discrimination and prediction of Klebsiella pneumoniae strains with carbapenem-sensitive, carbapenem-resistant, and polymyxin-resistant phenotypes. The implementation of CNN plus an attention mechanism makes the highest prediction accuracy at 99.46%, which confirms the diagnostic potential of the combination of SERS spectroscopy with the deep learning algorithm for antibacterial susceptibility testing in clinical settings.

Deep Learning and Antibiotic Resistance

Antibiotic resistance (AR) is a naturally occurring phenomenon with the capacity to render useless all known antibiotics in the fight against bacterial infections. Although bacterial resistance appeared before any human life form, this process has accelerated in the past years. Important causes of AR in modern times could be the over-prescription of antibiotics, the presence of faulty infection-prevention strategies, pollution in overcrowded areas, or the use of antibiotics in agriculture and farming, together with a decreased interest from the pharmaceutical industry in researching and testing new antibiotics. The last cause is primarily due to the high costs of developing antibiotics. The aim of the present review is to highlight the techniques that are being developed for the identification of new antibiotics to assist this lengthy process, using artificial intelligence (AI). AI can shorten the preclinical phase by rapidly generating many substances based on algorithms created by machine learning (ML) through techniques such as neural networks (NN) or deep learning (DL). Recently, a text mining system that incorporates DL algorithms was used to help and speed up the data curation process. Moreover, new and old methods are being used to identify new antibiotics, such as the combination of quantitative structure-activity relationship (QSAR) methods with ML or Raman spectroscopy and MALDI-TOF MS combined with NN, offering faster and easier interpretation of results. Thus, AI techniques are important additional tools for researchers and clinicians in the race for new methods of overcoming bacterial resistance.

Inflammation and fibrosis in the coal dust-exposed lung described by confocal Raman spectroscopy

Background

Coal workers' pneumoconiosis (CWP) is an occupational disease that severely damages the life and health of miners. However, little is known about the molecular and cellular mechanisms changes associated with lung inflammation and fibrosis induced by coal dust. As a non-destructive technique for measuring biological tissue, confocal Raman spectroscopy provides accurate molecular fingerprints of label-free tissues and cells. Here, the progression of lung inflammation and fibrosis in a murine model of CWP was evaluated using confocal Raman spectroscopy.

Methods

A mouse model of CWP was constructed and biochemical analysis in lungs exposed to coal dust after 1 month (CWP-1M) and 3 months (CWP-3M) vs control tissues (NS) were used by confocal Raman spectroscopy. H&E, immunohistochemical and collagen staining were used to evaluate the histopathology alterations in the lung tissues.

Results

The CWP murine model was successfully constructed, and the mouse lung tissues showed progression of inflammation and fibrosis, accompanied by changes in NF-κB, p53, Bax, and Ki67. Meanwhile, significant differences in Raman bands were observed among the different groups, particularly changes at 1,248, 1,448, 1,572, and 746 cm^-1. These changes were consistent with collagen, Ki67, and Bax levels in the CWP and NS groups.

Conclusion

Confocal Raman spectroscopy represented a novel approach to the identification of the biochemical changes in CWP lungs and provides potential biomarkers of inflammation and fibrosis.