Using random forest to predict antimicrobial minimum inhibitory concentrations of nontyphoidal Salmonella in Taiwan

Artificial intelligence in drug resistance management

This review highlights the application of artificial intelligence (AI), particularly deep learning and machine learning (ML), in managing antimicrobial resistance (AMR). Key findings demonstrate that AI models, such as Naïve Bayes, Decision Trees (DT), Random Forest (RF), Support Vector Machines (SVM), and Artificial Neural Networks (ANN), have significantly advanced the prediction of drug resistance patterns and the identification of novel antibiotics. These algorithms have effectively optimized antibiotic use, predicted resistance phenotypes, and identified new drug candidates. AI has also facilitated the detection of AMR-associated mutations, offering new insights into the spread of resistance and potential interventions. Despite data privacy and algorithm transparency challenges, AI presents a promising tool in combating AMR, with implications for improving patient outcomes, enhancing disease management, and addressing global public health concerns. However, realizing its full potential requires overcoming issues related to data scarcity, ethical considerations, and fostering interdisciplinary collaboration.

Quantitative prediction of disinfectant tolerance in Listeria monocytogenes using whole genome sequencing and machine learning

Listeria monocytogenes is a potentially severe disease-causing bacteria mainly transmitted through food. This pathogen is of great concern for public health and the food industry in particular. Many countries have implemented thorough regulations, and some have even set 'zero-tolerance' thresholds for particular food products to minimise the risk of L. monocytogenes outbreaks. This emphasises that proper sanitation of food processing plants is of utmost importance. Consequently, in recent years, there has been an increased interest in L. monocytogenes tolerance to disinfectants used in the food industry. Even though many studies are focusing on laboratory quantification of L. monocytogenes tolerance, the possibility of predictive models remains poorly studied. Within this study, we explore the prediction of tolerance and minimum inhibitory concentrations (MIC) using whole genome sequencing (WGS) and machine learning (ML). We used WGS data and MIC values to quaternary ammonium compound (QAC) disinfectants from 1649 L. monocytogenes isolates to train different ML predictors. Our study shows promising results for predicting tolerance to QAC disinfectants using WGS and machine learning. We were able to train high-performing ML classifiers to predict tolerance with balanced accuracy scores up to 0.97 ± 0.02. For the prediction of MIC values, we were able to train ML regressors with mean squared error as low as 0.07 ± 0.02. We also identified several new genes related to cell wall anchor domains, plasmids, and phages, putatively associated with disinfectant tolerance in L. monocytogenes. The findings of this study are a first step towards prediction of L. monocytogenes tolerance to QAC disinfectants used in the food industry. In the future, predictive models might be used to monitor disinfectant tolerance in food production and might support the conceptualisation of more nuanced sanitation programs.

Integrating sequencing methods with machine learning for antimicrobial susceptibility testing in pediatric infections: current advances and future insights

Antimicrobial resistance (AMR) presents a critical challenge in clinical settings, particularly among pediatric patients with life-threatening conditions such as sepsis, meningitis, and neonatal infections. The increasing prevalence of multi- and pan-resistant pathogens is strongly associated with adverse clinical outcomes. Recent technological advances in sequencing methods, including metagenomic next-generation sequencing (mNGS), Oxford Nanopore Technologies (ONT), and targeted sequencing (TS), have significantly enhanced the detection of both pathogens and their associated resistance genes. However, discrepancies between resistance gene detection and antimicrobial susceptibility testing (AST) often hinder the direct clinical application of sequencing results. These inconsistencies may arise from factors such as genetic mutations or variants in resistance genes, differences in the phenotypic expression of resistance, and the influence of environmental conditions on resistance levels, which can lead to variations in the observed resistance patterns. Machine learning (ML) provides a promising solution by integrating large-scale resistance data with sequencing outcomes, enabling more accurate predictions of pathogen drug susceptibility. This review explores the application of sequencing technologies and ML in the context of pediatric infections, with a focus on their potential to track the evolution of resistance genes and predict antibiotic susceptibility. The goal of this review is to promote the incorporation of ML-based predictions into clinical practice, thereby improving the management of AMR in pediatric populations.

Computer-Aided Discovery of Synergistic Drug-Nanoparticle Combinations for Enhanced Antimicrobial Activity

Antibiotic resistance is a critical global public health challenge driven by the limited discovery of antibiotics, the rapid evolution of resistance mechanisms, and persistent infections that compromise treatment efficacy. Combination therapies using antibiotics and nanoparticles (NPs) offer a promising solution, particularly against multidrug-resistant (MDR) bacteria. This study introduces an innovative approach to identifying synergistic drug-NP combinations with enhanced antimicrobial activity. To carry this out, we compiled two groups of data sets to predict the minimal concentration (MC) and zone of inhibition (ZOI) of various drug-NP combinations. CatBoost regression models achieved the best 10-fold cross-validation R2 scores of 0.86 and 0.77, respectively. We then adopted a machine learning (ML)-reinforced genetic algorithm (GA) to identify synergistic antimicrobial NPs. The proposed approach was first validated by reproducing the previous experimental results. As a proof of concept for discovering drug-NP combinations, Au NPs were identified as highly synergistic NPs when paired with chloramphenicol, achieving a minimum bactericidal concentration (MBC) of 71.74 ng/mL against Salmonella typhimurium with a fractional inhibitory concentration index of 6.23 × 10-3. These findings present an effective strategy for identifying synergistic drug-NP combinations, providing a promising approach to combating drug-resistant pathogens and advancing targeted antimicrobial therapies.

A novel multitask learning algorithm for tasks with distinct chemical space: zebrafish toxicity prediction as an example

Data scarcity is one of the most critical issues impeding the development of prediction models for chemical effects. Multitask learning algorithms leveraging knowledge from relevant tasks showed potential for dealing with tasks with limited data. However, current multitask methods mainly focus on learning from datasets whose task labels are available for most of the training samples. Since datasets were generated for different purposes with distinct chemical spaces, the conventional multitask learning methods may not be suitable. This study presents a novel multitask learning method MTForestNet that can deal with data scarcity problems and learn from tasks with distinct chemical space. The MTForestNet consists of nodes of random forest classifiers organized in the form of a progressive network, where each node represents a random forest model learned from a specific task. To demonstrate the effectiveness of the MTForestNet, 48 zebrafish toxicity datasets were collected and utilized as an example. Among them, two tasks are very different from other tasks with only 1.3% common chemicals shared with other tasks. In an independent test, MTForestNet with a high area under the receiver operating characteristic curve (AUC) value of 0.911 provided superior performance over compared single-task and multitask methods. The overall toxicity derived from the developed models of zebrafish toxicity is well correlated with the experimentally determined overall toxicity. In addition, the outputs from the developed models of zebrafish toxicity can be utilized as features to boost the prediction of developmental toxicity. The developed models are effective for predicting zebrafish toxicity and the proposed MTForestNet is expected to be useful for tasks with distinct chemical space that can be applied in other tasks.Scieific contributionA novel multitask learning algorithm MTForestNet was proposed to address the challenges of developing models using datasets with distinct chemical space that is a common issue of cheminformatics tasks. As an example, zebrafish toxicity prediction models were developed using the proposed MTForestNet which provide superior performance over conventional single-task and multitask learning methods. In addition, the developed zebrafish toxicity prediction models can reduce animal testing.

Machine learning and feature extraction for rapid antimicrobial resistance prediction of Acinetobacter baumannii from whole-genome sequencing data

Background

Whole-genome sequencing (WGS) has contributed significantly to advancements in machine learning methods for predicting antimicrobial resistance (AMR). However, the comparisons of different methods for AMR prediction without requiring prior knowledge of resistance remains to be conducted.

Methods

We aimed to predict the minimum inhibitory concentrations (MICs) of 13 antimicrobial agents against Acinetobacter baumannii using three machine learning algorithms (random forest, support vector machine, and XGBoost) combined with k-mer features extracted from WGS data.

Results

A cohort of 339 isolates was used for model construction. The average essential agreement and category agreement of the best models exceeded 90.90% (95%CI, 89.03-92.77%) and 95.29% (95%CI, 94.91-95.67%), respectively; the exceptions being levofloxacin, minocycline and imipenem. The very major error rates ranged from 0.0 to 5.71%. We applied feature selection pipelines to extract the top-ranked 11-mers to optimise training time and computing resources. This approach slightly improved the prediction performance and enabled us to obtain prediction results within 10 min. Notably, when employing these top-ranked 11-mers in an independent test dataset (120 isolates), we achieved an average accuracy of 0.96.

Conclusion

Our study is the first to demonstrate that AMR prediction for A. baumannii using machine learning methods based on k-mer features has competitive performance over traditional workflows; hence, sequence-based AMR prediction and its application could be further promoted. The k-mer-based workflow developed in this study demonstrated high recall/sensitivity and specificity, making it a dependable tool for MIC prediction in clinical settings.

Predicting Salmonella MIC and Deciphering Genomic Determinants of Antibiotic Resistance and Susceptibility

Salmonella spp., a leading cause of foodborne illness, is a formidable global menace due to escalating antimicrobial resistance (AMR). The evaluation of minimum inhibitory concentration (MIC) for antimicrobials is critical for characterizing AMR. The current whole genome sequencing (WGS)-based approaches for predicting MIC are hindered by both computational and feature identification constraints. We propose an innovative methodology called the "Genome Feature Extractor Pipeline" that integrates traditional machine learning (random forest, RF) with deep learning models (multilayer perceptron (MLP) and DeepLift) for WGS-based MIC prediction. We used a dataset from the National Antimicrobial Resistance Monitoring System (NARMS), comprising 4500 assembled genomes of nontyphoidal Salmonella, each annotated with MIC metadata for 15 antibiotics. Our pipeline involves the batch downloading of annotated genomes, the determination of feature importance using RF, Gini-index-based selection of crucial 10-mers, and their expansion to 20-mers. This is followed by an MLP network, with four hidden layers of 1024 neurons each, to predict MIC values. Using DeepLift, key 20-mers and associated genes influencing MIC are identified. The 10 most significant 20-mers for each antibiotic are listed, showcasing our ability to discern genomic features affecting Salmonella MIC prediction with enhanced precision. The methodology replaces binary indicators with k-mer counts, offering a more nuanced analysis. The combination of RF and MLP addresses the limitations of the existing WGS approach, providing a robust and efficient method for predicting MIC values in Salmonella that could potentially be applied to other pathogens.