Prediction of Antimicrobial Minimal Inhibitory Concentrations for Neisseria gonorrhoeae using Machine Learning Models

Predicting S. aureus antimicrobial resistance with interpretable genomic space maps

Increasing antimicrobial resistance (AMR) represents a global healthcare threat. To decrease the spread of AMR and associated mortality, methods for rapid selection of optimal antibiotic treatment are urgently needed. Machine learning (ML) models based on genomic data to predict resistant phenotypes can serve as a fast screening tool prior to phenotypic testing. Nonetheless, many existing ML methods lack interpretability. Therefore, we present a methodology for visualization of sequence space and AMR prediction based on the non-linear dimensionality reduction method - generative topographic mapping (GTM). This approach, applied to AMR data of >5000 S. aureus isolates retrieved from the PATRIC database, yielded GTM models with reasonable accuracy for all drugs (balanced accuracy values ≥0.75). The Generative Topographic Maps (GTMs) represent data in the form of illustrative maps of the genomic space and allow for antibiotic-wise comparison of resistant phenotypes. The maps were also found to be useful for the analysis of genetic determinants responsible for drug resistance. Overall, the GTM-based methodology is a useful tool for both the illustrative exploration of the genomic sequence space and AMR prediction.

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.

Comparative genomics reveals the correlations of stress response genes and bacteriophages in developing antibiotic resistance of Staphylococcus saprophyticus

IMPORTANCE

Staphylococcus saprophyticus is the second most common bacteria associated with urinary tract infections (UTIs) in women. The antimicrobial treatment regimen for uncomplicated UTI is normally nitrofurantoin, trimethoprim-sulfamethoxazole (TMP-SMX), or a fluoroquinolone without routine susceptibility testing of S. saprophyticus recovered from urine specimens. However, TMP-SMX-resistant S. saprophyticus has been detected recently in UTI patients, as well as in our cohort. Herein, we investigated the understudied resistance patterns of this pathogenic species by linking genomic antibiotic resistance gene (ARG) content to susceptibility phenotypes. We describe ARG associations with known and novel SCCmec configurations as well as phage elements in S. saprophyticus, which may serve as intervention or diagnostic targets to limit resistance transmission. Our analyses yielded a comprehensive database of phenotypic data associated with the ARG sequence in clinical S. saprophyticus isolates, which will be crucial for resistance surveillance and prediction to enable precise diagnosis and effective treatment of S. saprophyticus UTIs.

Machine learning to predict ceftriaxone resistance using single nucleotide polymorphisms within a global database of Neisseria gonorrhoeae genomes

IMPORTANCE

Antimicrobial resistance in Neisseria gonorrhoeae is an urgent global health issue. The objectives of the study were to use a global collection of 12,936 N. gonorrhoeae genomes from the PathogenWatch database to evaluate different machine learning models to predict ceftriaxone susceptibility/decreased susceptibility using 97 mutations known to be associated with ceftriaxone resistance. We found the random forest classifier model had the highest performance. The analysis also reported the relative contributions of different mutations within the ML model predictions, allowing for the identification of the mutations with the highest importance for ceftriaxone resistance. A machine learning model retrained with the top five mutations performed similarly to the model using all 97 mutations. These results could aid in the development of molecular tests to detect resistance to ceftriaxone in N. gonorrhoeae. Moreover, this approach could be applied to building and evaluating machine learning models for predicting antimicrobial resistance in other pathogens.

Antibiotic resistance genes prevalence prediction and interpretation in beaches affected by urban wastewater discharge

Background

The annual death toll of over 1.2 million worldwide is attributed to infections caused by resistant bacteria, driven by the significant impact of antibiotic misuse and overuse in spreading these bacteria and their associated antibiotic resistance genes (ARGs). While limited data suggest the presence of ARGs in beach environments, efficient prediction tools are needed for monitoring and detecting ARGs to ensure public health safety. This study aims to develop interpretable machine learning methods for predicting ARGs in beach waters, addressing the challenge of black-box models and enhancing our understanding of their internal mechanisms.

Methods

In this study, we systematically collected beach water samples and subsequently isolated bacteria from these samples using various differential and selective media supplemented with different antibiotics. Resistance profiles of bacteria were determined by using Kirby-Bauer disk diffusion method. Further, ARGs were enumerated by using the quantitative polymerase chain reaction (qPCR) to detect and quantify ARGs. The obtained qPCR data and hydro-meteorological were used to create an ML model with high prediction performance and we further used two explainable artificial intelligence (xAI) model-agnostic interpretation methods to describe the internal behavior of ML model.

Results

Using qPCR, we detected blaCTX-M, blaNDM, blaCMY, blaOXA, blatetX, blasul1, and blaaac(6'-Ib-cr) in the beach waters. Further, we developed ML prediction models for blaaac(6'-Ib-cr), blasul1, and blatetX using the hydro-metrological and qPCR-derived data and the models demonstrated strong performance, with R2 values of 0.957, 0.997, and 0.976, respectively.

Conclusions

Our findings show that environmental factors, such as water temperature, precipitation, and tide, are among the important predictors of the abundance of resistance genes at beaches.

Bioinformatics analysis and prediction of Alzheimer's disease and alcohol dependence based on Ferroptosis-related genes

Background

Alzheimer's disease (AD) is a neurodegenerative disease whose origins have not been universally accepted. Numerous studies have demonstrated the relationship between AD and alcohol dependence; however, few studies have combined the origins of AD, alcohol dependence, and programmed cell death (PCD) to analyze the mechanistic relationship between the development of this pair of diseases. We demonstrated in previous studies the relationship between psychiatric disorders and PCD, and in the same concerning neurodegeneration-related AD, we found an interesting link with the Ferroptosis pathway. In the present study, we explored the bioinformatic interactions between AD, alcohol dependence, and Ferroptosis and tried to elucidate and predict the development of AD from this aspect.

Methods

We selected the Alzheimer's disease dataset GSE118553 and alcohol dependence dataset GSE44456 from the Gene Expression Omnibus (GEO) database. Ferroptosis-related genes were gathered through Gene Set Enrichment Analysis (GSEA), Kyoto Encyclopedia of Genes and Genomes (KEGG), and relevant literature, resulting in a total of 88 related genes. For the AD and alcohol dependence datasets, we conducted Limma analysis to identify differentially expressed genes (DEGs) and performed functional enrichment analysis on the intersection set. Furthermore, we used ferroptosis-related genes and the DEGs to perform machine learning crossover analysis, employing Least Absolute Shrinkage and Selection Operator (LASSO) regression to identify candidate immune-related central genes. This analysis was also used to construct protein-protein interaction networks (PPI) and artificial neural networks (ANN), as well as to plot receiver operating characteristic (ROC) curves for diagnosing AD and alcohol dependence. We analyzed immune cell infiltration to explore the role of immune cell dysregulation in AD. Subsequently, we conducted consensus clustering analysis of AD using three relevant candidate gene models and examined the immune microenvironment and functional pathways between different subgroups. Finally, we generated a network of gene-gene interactions and miRNA-gene interactions using Networkanalyst.

Results

The crossover of AD and alcohol dependence DEG contains 278 genes, and functional enrichment analysis showed that both AD and alcohol dependence were strongly correlated with Ferroptosis, and then crossed them with Ferroptosis-related genes to obtain seven genes. Three candidate genes were finally identified by machine learning to build a diagnostic prediction model. After validation by ANN and PPI analysis, ROC curves were plotted to assess the diagnostic value of AD and alcohol dependence. The results showed a high diagnostic value of the predictive model. In the immune infiltration analysis, functional metabolism and immune microenvironment of AD patients were significantly associated with Ferroptosis. Finally, analysis of target genes and miRNA-gene interaction networks showed that hsa-mir-34a-5p and has-mir-106b-5p could simultaneously regulate the expression of both CYBB and ACSL4.

Conclusion

We obtained a diagnostic prediction model with good effect by comprehensive analysis, and validation of ROC in AD and alcohol dependence data sets showed good diagnostic, predictive value for both AD (AUC 0. 75, CI 0.91-0.60), and alcohol dependence (AUC 0.81, CI 0.95-0.68). In the consensus clustering grouping, we identified variability in the metabolic and immune microenvironment between subgroups as a likely cause of the different prognosis, which was all related to Ferroptosis function. Finally, we discovered that hsa-mir-34a-5p and has-mir-106b-5p could simultaneously regulate the expression of both CYBB and ACSL4.

Predictive model, miRNA-TF network, related subgroup identification and drug prediction of ischemic stroke complicated with mental disorders based on genes related to gut microbiome

Background

Patients with comorbid schizophrenia, depression, drug use, and multiple psychiatric diagnoses have a greater risk of carotid revascularization following stroke. The gut microbiome (GM) plays a crucial role in the attack of mental illness and IS, which may become an index for the diagnosis of IS. A genomic study of the genetic commonalities between SC and IS, as well as its mediated pathways and immune infiltration, will be conducted to determine how schizophrenia contributes to the high prevalence of IS. According to our study, this could be an indicator of ischemic stroke development.

Methods

We selected two datasets of IS from the Gene Expression Omnibus (GEO), one for training and the other for the verification group. Five genes related to mental disorders and GM were extracted from Gene cards and other databases. Linear models for microarray data (Limma) analysis was utilized to identify differentially expressed genes (DEGs) and perform functional enrichment analysis. It was also used to conduct machine learning exercises such as random forest and regression to identify the best candidate for immune-related central genes. Protein-protein interaction (PPI) network and artificial neural network (ANN) were established for verification. The receiver operating characteristic (ROC) curve was drawn for the diagnosis of IS, and the diagnostic model was verified by qRT-PCR. Further immune cell infiltration analysis was performed to study the IS immune cell imbalance. We also performed consensus clustering (CC) to analyze the expression of candidate models under different subtypes. Finally, miRNA, transcription factors (TFs), and drugs related to candidate genes were collected through the Network analyst online platform.

Results

Through comprehensive analysis, a diagnostic prediction model with good effect was obtained. Both the training group (AUC 0.82, CI 0.93-0.71) and the verification group (AUC 0.81, CI 0.90-0.72) had a good phenotype in the qRT-PCR test. And in verification group 2 we validated between the two groups with and without carotid-related ischemic cerebrovascular events (AUC 0.87, CI 1-0.64). Furthermore, we investigated cytokines in both GSEA and immune infiltration and verified cytokine-related responses by flow cytometry, particularly IL-6, which played an important role in IS occurrence and progression. Therefore, we speculate that mental illness may affect the development of IS in B cells and IL-6 in T cells. MiRNA (hsa-mir-129-2-3p, has-mir-335-5p, and has-mir-16-5p) and TFs (CREB1, FOXL1), which may be related to IS, were obtained.

Conclusion

Through comprehensive analysis, a diagnostic prediction model with good effect was obtained. Both the training group (AUC 0.82, CI 0.93-0.71) and the verification group (AUC 0.81, CI 0.90-0.72) had a good phenotype in the qRT-PCR test. And in verification group 2 we validated between the two groups with and without carotid-related ischemic cerebrovascular events (AUC 0.87, CI 1-0.64). MiRNA (hsa-mir-129-2-3p, has-mir-335-5p, and has-mir-16-5p) and TFs (CREB1, FOXL1), which may be related to IS, were obtained.

Artificial Intelligence for Antimicrobial Resistance Prediction: Challenges and Opportunities towards Practical Implementation

Antimicrobial resistance (AMR) is emerging as a potential threat to many lives worldwide. It is very important to understand and apply effective strategies to counter the impact of AMR and its mutation from a medical treatment point of view. The intersection of artificial intelligence (AI), especially deep learning/machine learning, has led to a new direction in antimicrobial identification. Furthermore, presently, the availability of huge amounts of data from multiple sources has made it more effective to use these artificial intelligence techniques to identify interesting insights into AMR genes such as new genes, mutations, drug identification, conditions favorable to spread, and so on. Therefore, this paper presents a review of state-of-the-art challenges and opportunities. These include interesting input features posing challenges in use, state-of-the-art deep-learning/machine-learning models for robustness and high accuracy, challenges, and prospects to apply these techniques for practical purposes. The paper concludes with the encouragement to apply AI to the AMR sector with the intention of practical diagnosis and treatment, since presently most studies are at early stages with minimal application in the practice of diagnosis and treatment of disease.

Prediction and interpretation of antibiotic-resistance genes occurrence at recreational beaches using machine learning models

Antibiotic-resistant bacteria and antibiotic resistance genes (ARGs) are pollutants of worldwide concern that seriously threaten public health and ecosystems. Machine learning (ML) prediction models have been applied to predict ARGs in beach waters. However, the existing studies were conducted at a single location and had low prediction performance. Moreover, ML models are "black boxes" that do not reveal their predictions' internal nuances and mechanisms. This lack of transparency and trust can result in serious consequences when using these models in high-stakes decisions. In this study, we developed a gradient boosted regression tree based (GBRT) ML model and then described its behavior using six explainable artificial intelligence (XAI) model-agnostic explanation methods. We used hydro-meteorological and qPCR data from the beaches in South Korea and Pakistan and developed ML prediction models for aac (6'-lb-cr), sul1, and tetX with 10-fold time-blocked cross-validation performances of 4.9, 2.06 and 4.4 root mean squared logarithmic error, respectively. We then analyzed the local and global behavior of the developed ML model using four interpretation methods. The developed ML models showed that water temperature, precipitation and tide are the most important predictors for prediction of ARGs at recreational beaches. We show that the model-agnostic interpretation methods not only explain the behavior of the ML model but also provide insights into the behavior of the ML model under new unseen conditions. Moreover, these post-processing techniques can be a debugging tool for ML-based modeling.

Machine learning models for Neisseria gonorrhoeae antimicrobial susceptibility tests

Neisseria gonorrhoeae is an urgent public health threat due to the emergence of antibiotic resistance. As most isolates in the United States are susceptible to at least one antibiotic, rapid molecular antimicrobial susceptibility tests (ASTs) would offer the opportunity to tailor antibiotic therapy, thereby expanding treatment options. With genome sequence and antibiotic resistance phenotype data for nearly 20,000 clinical N. gonorrhoeae isolates now available, there is an opportunity to use statistical methods to develop sequence-based diagnostics that predict antibiotic susceptibility from genotype. N. gonorrhoeae, therefore, provides a useful example illustrating how to apply machine learning models to aid in the design of sequence-based ASTs. We present an overview of this framework, which begins with establishing the assay technology, the performance criteria, the population in which the diagnostic will be used, and the clinical goals, and extends to the choices that must be made to arrive at a set of features with the desired properties for predicting susceptibility phenotype from genotype. While we focus on the example of N. gonorrhoeae, the framework generalizes to other organisms for which large-scale genotype and antibiotic resistance data can be combined to aid in diagnostics development.

Schizophrenia and cell senescence candidate genes screening, machine learning, diagnostic models, and drug prediction

Background

Schizophrenia (SC) is one of the most common psychiatric diseases. Its potential pathogenic genes and effective treatment methods are still unclear. Cell senescence has been confirmed in mental diseases. A link exists between cellular senescence and immunity, and immune-related problems affect suicide rates in individuals suffering from schizophrenia. Therefore, the aims of this study were to identify candidate genes based on cell senescence that can affect the diagnosis and treatment of schizophrenia.

Methods

Two data sets of schizophrenia were provided by the Gene Expression Omnibus (GEO) database, one was taken as training and the other as a validation group. The genes related to cell senescence were obtained from the CellAge database. DEGs were identified using the Limma package and weighted gene co-expression network analysis (WGCNA). The function enrichment analysis was conducted, followed by machine learning-based identification for least absolute shrinking and selection operators (LASSO) regression. Random Forest were used to identify candidate immune-related central genes and establish artificial neural networks for verification of the candidate genes. The receiver operating characteristic curve (ROC curve) was used for the diagnosis of schizophrenia. Immune cell infiltrates were constructed to study immune cell dysregulation in schizophrenia, and relevant drugs with candidate genes were collected from the DrugBank database.

Results

Thirteen co-expression modules were screened for schizophrenia, of which 124 were the most relevant genes.There were 23 intersected genes of schizophrenia (including DEGs and the cellular senescence-related genes), and through machine learning six candidate genes were finally screened out. The diagnostic value was evaluated using the ROC curve data. Based on these results it was confirmed that these candidate genes have high diagnostic value.Two drugs related to candidate genes, Fostamatinib and Ritodine, were collected from the DrugBanks database.

Conclusion

Six potential candidate genes (SFN, KDM5B, MYLK, IRF3, IRF7, and ID1) had been identified, all of which had diagnostic significance. Fostamatinib might be a drug choice for patients with schizophrenia to develop immune thrombocytopenic purpura (ITP) after treatment, providing effective evidence for the pathogenesis and drug treatment of schizophrenia.

Application of Decision-Tree-Based Machine Learning Algorithms for Prediction of Antimicrobial Resistance

Timely and efficacious antibiotic treatment depends on precise and quick in silico antimicrobial-resistance predictions. Limited treatment choices due to antimicrobial resistance (AMR) highlight the necessity to optimize the available diagnostics. AMR can be explicitly anticipated on the basis of genome sequence. In this study, we used transcriptomes of 410 multidrug-resistant isolates of Pseudomonas aeruginosa. We trained 10 machine learning (ML) classifiers on the basis of data on gene expression (GEXP) information and generated predictive models for meropenem, ciprofloxacin, and ceftazidime drugs. Among all the used ML models, four models showed high F1-score, accuracy, precision, and specificity compared with the other models. However, RandomForestClassifier showed a moderate F1-score (0.6), precision (0.61), and specificity (0.625) for ciprofloxacin. In the case of ceftazidime, RidgeClassifier performed well and showed F1-score (0.652), precision (0.654), and specificity (0.652) values. For meropenem, KNeighborsClassifier exhibited moderate F1-score (0.629), precision (0.629), and specificity (0.629). Among these three antibiotics, GEXP data on meropenem and ceftazidime improved diagnostic performance. The findings will pave the way for the establishment of a resistance profiling tool that can predict AMR on the basis of transcriptomic markers.

Machine learning analysis on the impacts of COVID-19 on India's renewable energy transitions and air quality

India is severely affected by the COVID-19 pandemic and is facing an unprecedented public health emergency. While the country's immediate measures focus on combating the coronavirus spread, it is important to investigate the impacts of the current crisis on India's renewable energy transition and air quality. India's economic slowdown is mainly compounded by the collapse of global oil prices and the erosion of global energy demand. A clean energy transition is a key step in enabling the integration of energy and climate. Millions in India are affected owing to fossil fuel pollution and the increasing climate heating that has led to inconceivable health impacts. This paper attempts to study the impact of COVID-19 on India's climate and renewable energy transitions through machine learning algorithms. India is observing a massive collapse in energy demand during the lockdown as its coal generation is suffering the worst part of the ongoing pandemic. During this current COVID-19 crisis, the renewable energy sector benefits from its competitive cost and the Indian government's must-run status to run generators based on renewable energy sources. In contrast to fossil fuel-based power plants, renewable energy sources are not exposed to the same supply chain disruptions in this current pandemic situation. India has the definite potential to surprise the global community and contribute to cost-effective decarbonization. Moreover, the country has a good chance of building more flexibility into the renewable energy sector to avoid an unstable future.