Accuracy of Different Bioinformatics Methods in Detecting Antibiotic Resistance and Virulence Factors from Staphylococcus aureus Whole-Genome Sequences

From Data to Decisions: Leveraging Artificial Intelligence and Machine Learning in Combating Antimicrobial Resistance - a Comprehensive Review

The emergence of drug-resistant bacteria poses a significant challenge to modern medicine. In response, Artificial Intelligence (AI) and Machine Learning (ML) algorithms have emerged as powerful tools for combating antimicrobial resistance (AMR). This review aims to explore the role of AI/ML in AMR management, with a focus on identifying pathogens, understanding resistance patterns, predicting treatment outcomes, and discovering new antibiotic agents. Recent advancements in AI/ML have enabled the efficient analysis of large datasets, facilitating the reliable prediction of AMR trends and treatment responses with minimal human intervention. ML algorithms can analyze genomic data to identify genetic markers associated with antibiotic resistance, enabling the development of targeted treatment strategies. Additionally, AI/ML techniques show promise in optimizing drug administration and developing alternatives to traditional antibiotics. By analyzing patient data and clinical outcomes, these technologies can assist healthcare providers in diagnosing infections, evaluating their severity, and selecting appropriate antimicrobial therapies. While integration of AI/ML in clinical settings is still in its infancy, advancements in data quality and algorithm development suggest that widespread clinical adoption is forthcoming. In conclusion, AI/ML holds significant promise for improving AMR management and treatment outcome.

Prediction of antimicrobial resistance in Klebsiella pneumoniae using genomic and metagenomic next-generation sequencing data

OBJECTIVES

Klebsiella pneumoniae is a significant pathogen with increasing resistance and high mortality rates. Conventional antibiotic susceptibility testing methods are time-consuming. Next-generation sequencing has shown promise for predicting antimicrobial resistance (AMR). This study aims to develop prediction models using whole-genome sequencing data and assess their feasibility with metagenomic next-generation sequencing data from clinical samples.

METHODS

On the basis of 4170 K. pneumoniae genomes, the main genetic characteristics associated with AMR were identified using a LASSO regression model. Consequently, the prediction model was established, validated and optimized using clinical isolate read simulation sequences. To evaluate the efficacy of the model, clinical specimens were collected.

RESULTS

Four predictive models for amikacin, ciprofloxacin, levofloxacin and piperacillin/tazobactam, initially had positive predictive values (PPVs) of 90%, 85%, 84% and 94%, respectively, when they were originally constructed. When applied to clinical specimens, their PPVs increased to 96%, 96%, 95% and 100%, respectively. Meanwhile, there were negative predictive values (NPVs) of 100% for ciprofloxacin and levofloxacin, and 'not applicable' (NA) for amikacin and piperacillin/tazobactam. Our method achieved antibacterial phenotype classification accuracy rates of 96.08% for amikacin, 96.15% for ciprofloxacin, 95.31% for levofloxacin and 100% for piperacillin/tazobactam. The sequence-based prediction antibiotic susceptibility testing (AST) reported results in an average time of 19.5 h, compared with the 67.9 h needed for culture-based AST, resulting in a significant reduction of 48.4 h.

CONCLUSIONS

These preliminary results demonstrated that the performance of prediction model for a clinically significant antimicrobial-species pair was comparable to that of phenotypic methods, thereby encouraging the expansion of sequence-based susceptibility prediction and its clinical validation and application.

Antimicrobial Resistance Genes in Respiratory Bacteria from Weaned Dairy Heifers

Bovine respiratory disease (BRD) is the leading cause of mortality and antimicrobial drug (AMD) use in weaned dairy heifers. Limited information is available regarding antimicrobial resistance (AMR) in respiratory bacteria in this population. This study determined AMR gene presence in 326 respiratory isolates (Pasteurella multocida, Mannheimia haemolytica, and Histophilus somni) from weaned dairy heifers using whole genome sequencing. Concordance between AMR genotype and phenotype was determined. Twenty-six AMR genes for 8 broad classes of AMD were identified. The most prevalent, medically important AMD classes used in calf rearing, to which these genes predict AMR among study isolates were tetracycline (95%), aminoglycoside (94%), sulfonamide (94%), beta-lactam (77%), phenicol (50%), and macrolide (44%). The co-occurrence of AMR genes within an isolate was common; the largest cluster of gene co-occurrence encodes AMR to phenicol, macrolide, elfamycin, β-lactam (cephalosporin, penam cephamycin), aminoglycoside, tetracycline, and sulfonamide class AMD. Concordance between genotype and phenotype varied (Matthew's Correlation Coefficient ranged from -0.57 to 1) by bacterial species, gene, and AMD tested, and was particularly poor for fluoroquinolones (no AMR genes detected) and ceftiofur (no phenotypic AMR classified while AMR genes present). These findings suggest a high genetic potential for AMR in weaned dairy heifers; preventing BRD and decreasing AMD reliance may be important in this population.

Historic methicillin-resistant Staphylococcus aureus: expanding current knowledge using molecular epidemiological characterization of a Swiss legacy collection

BACKGROUND

Few methicillin-resistant Staphylococcus aureus (MRSA) from the early years of its global emergence have been sequenced. Knowledge about evolutionary factors promoting the success of specific MRSA multi-locus sequence types (MLSTs) remains scarce. We aimed to characterize a legacy MRSA collection isolated from 1965 to 1987 and compare it against publicly available international and local genomes.

METHODS

We accessed 451 historic (1965-1987) MRSA isolates stored in the Culture Collection of Switzerland, mostly collected from the Zurich region. We determined phenotypic antimicrobial resistance (AMR) and performed whole genome sequencing (WGS) using Illumina short-read sequencing on all isolates and long-read sequencing on a selection with Oxford Nanopore Technology. For context, we included 103 publicly available international assemblies from 1960 to 1992 and sequenced 1207 modern Swiss MRSA isolates from 2007 to 2022. We analyzed the core genome (cg)MLST and predicted SCCmec cassette types, AMR, and virulence genes.

RESULTS

Among the 451 historic Swiss MRSA isolates, we found 17 sequence types (STs) of which 11 have been previously described. Two STs were novel combinations of known loci and six isolates carried previously unsubmitted MLST alleles, representing five new STs (ST7843, ST7844, ST7837, ST7839, and ST7842). Most isolates (83% 376/451) represented ST247-MRSA-I isolated in the 1960s, followed by ST7844 (6% 25/451), a novel single locus variant (SLV) of ST239. Analysis by cgMLST indicated that isolates belonging to ST7844-MRSA-III cluster within the diversity of ST239-MRSA-III. Early MRSA were predominantly from clonal complex (CC)8. From 1980 to the end of the twentieth century, we observed that CC22 and CC5 as well as CC8 were present, both locally and internationally.

CONCLUSIONS

The combined analysis of 1761 historic and contemporary MRSA isolates across more than 50 years uncovered novel STs and allowed us a glimpse into the lineage flux between Swiss-German and international MRSA across time.

Current and Future Technologies for the Detection of Antibiotic-Resistant Bacteria

Antibiotic resistance is a global public health concern, posing a significant threat to the effectiveness of antibiotics in treating bacterial infections. The accurate and timely detection of antibiotic-resistant bacteria is crucial for implementing appropriate treatment strategies and preventing the spread of resistant strains. This manuscript provides an overview of the current and emerging technologies used for the detection of antibiotic-resistant bacteria. We discuss traditional culture-based methods, molecular techniques, and innovative approaches, highlighting their advantages, limitations, and potential future applications. By understanding the strengths and limitations of these technologies, researchers and healthcare professionals can make informed decisions in combating antibiotic resistance and improving patient outcomes.

Functional network analysis identifies multiple virulence and antibiotic resistance targets in Stenotrophomonas maltophilia

Stenotrophomonas maltophilia, an emerging multidrug-resistant opportunistic bacterium in humans is of major concern for immunocompromised individuals for causing pneumonia and bloodborne infections. This bacterial pathogen is associated with a considerable fatality/case ratio, with up to 100%, when presented as hemorrhagic fever. It is resistant to commonly used drugs as well as to antibiotic combinations. In-silico based functional network analysis is a key approach to get novel insights into virulence and resistance in pathogenic organisms. This study included the protein-protein interaction (PPI) network analysis of 150 specific genes identified for antibiotic resistance mechanism and virulence pathways. Eight proteins, namely, PilL, FliA, Smlt2260, Smlt2267, CheW, Smlt2318, CheZ, and FliM were identified as hub proteins. Further docking studies of 58 selected phytochemicals were performed against the identified hub proteins. Deoxytubulosine and corosolic acid were found to be potent inhibitors of hub proteins of pathogenic S. maltophilia based on protein-ligand interactive study. Further pharmacophore studies are warranted with these molecules to develop them as novel antibiotics against S. maltophilia.

Identification of Glucocorticoid Receptor Target Genes That Potentially Inhibit Collagen Synthesis in Human Dermal Fibroblasts

Over several decades, excess glucocorticoids (GCs) of endogenous or exogenous origin have been recognized to significantly inhibit collagen synthesis and accelerate skin aging. However, little is known regarding their molecular mechanisms. We hypothesized that the action of GCs on collagen production is at least partially through the glucocorticoid receptor (GR) and its target genes, and therefore aimed to identify GR target genes that potentially inhibit collagen synthesis in Hs68 human dermal fibroblasts. We first confirmed that dexamethasone, a synthetic GC, induced canonical GR signaling in dermal fibroblasts. We then collected 108 candidates for GR target genes reported in previous studies on GR target genes and verified that 17 genes were transcriptionally upregulated in dexamethasone-treated dermal fibroblasts. Subsequently, by individual knockdown of the 17 genes, we identified that six genes, AT-rich interaction domain 5B, FK506 binding protein 5, lysyl oxidase, methylenetetrahydrofolate dehydrogenase (NADP + dependent) 2, zinc finger protein 36, and zinc fingers and homeoboxes 3, are potentially involved in GC-mediated inhibition of collagen synthesis. The present study sheds light on the molecular mechanisms of GC-mediated skin aging and provides a basis for further research on the biological characteristics of individual GR target genes.

Applied genomics for identification of virulent biothreats and for disease outbreak surveillance

Fortifying our preparedness to cope with biological threats by identifying and targeting virulence factors may be a preventative strategy for curtailing infectious disease outbreak. Virulence factors evoke successful pathogenic invasion, and the science and technology of genomics offers a way of identifying them, their agents and evolutionary ancestor. Genomics offers the possibility of deciphering if the release of a pathogen was intentional or natural by observing sequence and annotated data of the causative agent, and evidence of genetic engineering such as cloned vectors at restriction sites. However, to leverage and maximise the application of genomics to strengthen global interception system for real-time biothreat diagnostics, a complete genomic library of pathogenic and non-pathogenic agents will create a robust reference assembly that can be used to screen, characterise, track and trace new and existing strains. Encouraging ethical research sequencing pathogens found in animals and the environment, as well as creating a global space for collaboration will lead to effective global regulation and biosurveillance.

Advantages of Whole Genome Sequencing in Mitigating the Helicobacter pylori Antimicrobial Resistance Problem

Helicobacter pylori antimicrobial resistance is a critical public health issue. Typically, antimicrobial resistance epidemiology reports include only the antimicrobial susceptibility test results for H. pylori. However, this phenotypic approach is less capable of answering queries related to resistance mechanisms and specific mutations found in particular global regions. Whole genome sequencing can help address these two questions while still offering quality control and is routinely validated against AST standards. A comprehensive understanding of the mechanisms of resistance should improve H. pylori eradication efforts and prevent gastric cancer.

Benchmarking of two bioinformatic workflows for the analysis of whole-genome sequenced Staphylococcus aureus collected from patients with suspected sepsis

BACKGROUND

The rapidly growing area of sequencing technologies, and more specifically bacterial whole-genome sequencing, could offer applications in clinical microbiology, including species identification of bacteria, prediction of genetic antibiotic susceptibility and virulence genes simultaneously. To accomplish the aforementioned points, the commercial cloud-based platform, 1928 platform (1928 Diagnostics, Gothenburg, Sweden) was benchmarked against an in-house developed bioinformatic pipeline as well as to reference methods in the clinical laboratory.

METHODS

Whole-genome sequencing data retrieved from 264 Staphylococcus aureus isolates using the Illumina HiSeq X next-generation sequencing technology was used. The S. aureus isolates were collected during a prospective observational study of community-onset severe sepsis and septic shock in adults at Skaraborg Hospital, in the western region of Sweden. The collected isolates were characterized according to accredited laboratory methods i.e., species identification by MALDI-TOF MS analysis and phenotypic antibiotic susceptibility testing (AST) by following the EUCAST guidelines. Concordance between laboratory methods and bioinformatic tools, as well as concordance between the bioinformatic tools was assessed by calculating the percent of agreement.

RESULTS

There was an overall high agreement between predicted genotypic AST and phenotypic AST results, 98.0% (989/1006, 95% CI 97.3-99.0). Nevertheless, the 1928 platform delivered predicted genotypic AST results with lower very major error rates but somewhat higher major error rates compared to the in-house pipeline. There were differences in processing times i.e., minutes versus hours, where the 1928 platform delivered the results faster. Furthermore, the bioinformatic workflows showed overall 99.4% (1267/1275, 95% CI 98.7-99.7) agreement in genetic prediction of the virulence gene characteristics and overall 97.9% (231/236, 95% CI 95.0-99.2%) agreement in predicting the sequence types (ST) of the S. aureus isolates.

CONCLUSIONS

Altogether, the benchmarking disclosed that both bioinformatic workflows are able to deliver results with high accuracy aiding diagnostics of severe infections caused by S. aureus. It also illustrates the need of international agreement on quality control and metrics to facilitate standardization of analytical approaches for whole-genome sequencing based predictions.

Genomic characterization and assessment of pathogenic potential of Legionella spp. isolates from environmental monitoring

Several species in the genus Legionella are known to cause an acute pneumonia when the aerosols containing the bacteria from man-made water systems are inhaled. The disease is usually caused by Legionella pneumophila, but other species have been implicated in the infection. The disease is frequently manifested as an outbreak, which means several people are affected when exposed to the common source of Legionella contamination. Therefor environmental surveillance which includes isolation and identification of Legionella is performed routinely. However, usually no molecular or genome-based methods are employed in further characterization of the isolates during routine environmental monitoring. During several years of such monitoring, isolates from different geographical locations were collected and 39 of them were sequenced by hybrid de novo approach utilizing short and long sequencing reads. In addition, the isolates were typed by standard culture and MALDI-TOF method. The sequencing reads were assembled and annotated to produce high-quality genomes. By employing discriminatory genome typing, four potential new species in the Legionella genus were identified, which are yet to be biochemically and morphologically characterized. Moreover, functional annotations concerning virulence and antimicrobial resistance were performed on the sequenced genomes. The study contributes to the knowledge on little-known non-pneumophila species present in man-made water systems and establishes support for future genetic relatedness studies as well as understanding of their pathogenic potential.

Next Generation and Other Sequencing Technologies in Diagnostic Microbiology and Infectious Diseases

Next-generation sequencing (NGS) technologies have become increasingly available for use in the clinical microbiology diagnostic environment. There are three main applications of these technologies in the clinical microbiology laboratory: whole genome sequencing (WGS), targeted metagenomics sequencing and shotgun metagenomics sequencing. These applications are being utilized for initial identification of pathogenic organisms, the detection of antimicrobial resistance mechanisms and for epidemiologic tracking of organisms within and outside hospital systems. In this review, we analyze these three applications and provide a comprehensive summary of how these applications are currently being used in public health, basic research, and clinical microbiology laboratory environments. In the public health arena, WGS is being used to identify and epidemiologically track food borne outbreaks and disease surveillance. In clinical hospital systems, WGS is used to identify multi-drug-resistant nosocomial infections and track the transmission of these organisms. In addition, we examine how metagenomics sequencing approaches (targeted and shotgun) are being used to circumvent the traditional and biased microbiology culture methods to identify potential pathogens directly from specimens. We also expand on the important factors to consider when implementing these technologies, and what is possible for these technologies in infectious disease diagnosis in the next 5 years.

Leveraging big data bioinformatics approaches to extract knowledge from Staphylococcus aureus public omics data

Staphylococcus aureus is a notorious pathogen posing challenges in the medical industry due to drug resistance and biofilm formation. The horizon of knowledge on S. aureus pathogenesis has expanded with the advancement of data-driven bioinformatics techniques. Mining information from sequenced genomes and their expression data is an economic approach that alleviates wastage of resources and redundancy in experiments. The current review covers how big data bioinformatics has been used in the analysis of S. aureus from publicly available -omics data to uncover mechanisms of infection and inhibition. Particularly, advances in the past two decades in biomarker discovery, host responses, phenotype identification, consolidation of information, and drug development are discussed highlighting the challenges and shortcomings. Overall, the review summarizes the diverse aspects of scrupulous re-analysis of S. aureus proteomic and transcriptomic expression datasets retrieved from public repositories in terms of the efforts taken, benefits offered, and follow-up actions. The detailed review thus serves as a reference and aid for (i) Computational biologists by briefing the approaches utilized for bacterial omics re-analysis concerning S. aureus and (ii) Experimental biologists by elucidating the potential of bioinformatics in biological research to generate reliable postulates in a prompt and economical manner.

A Practical Approach for Predicting Antimicrobial Phenotype Resistance in Staphylococcus aureus Through Machine Learning Analysis of Genome Data

With the reduction in sequencing price and acceleration of sequencing speed, it is particularly important to directly link the genotype and phenotype of bacteria. Here, we firstly predicted the minimum inhibitory concentrations of ten antimicrobial agents for Staphylococcus aureus using 466 isolates by directly extracting k-mer from whole genome sequencing data combined with three machine learning algorithms: random forest, support vector machine, and XGBoost. Considering one two-fold dilution, the essential agreement and the category agreement could reach >85% and >90% for most antimicrobial agents. For clindamycin, cefoxitin and trimethoprim-sulfamethoxazole, the essential agreement and the category agreement could reach >91% and >93%, providing important information for clinical treatment. The successful prediction of cefoxitin resistance showed that the model could identify methicillin-resistant S. aureus. The results suggest that small datasets available in large hospitals could bypass the existing basic research and known antimicrobial resistance genes and accurately predict the bacterial phenotype.

A roadmap for the generation of benchmarking resources for antimicrobial resistance detection using next generation sequencing

Next Generation Sequencing technologies significantly impact the field of Antimicrobial Resistance (AMR) detection and monitoring, with immediate uses in diagnosis and risk assessment. For this application and in general, considerable challenges remain in demonstrating sufficient trust to act upon the meaningful information produced from raw data, partly because of the reliance on bioinformatics pipelines, which can produce different results and therefore lead to different interpretations. With the constant evolution of the field, it is difficult to identify, harmonise and recommend specific methods for large-scale implementations over time. In this article, we propose to address this challenge through establishing a transparent, performance-based, evaluation approach to provide flexibility in the bioinformatics tools of choice, while demonstrating proficiency in meeting common performance standards. The approach is two-fold: first, a community-driven effort to establish and maintain “live” (dynamic) benchmarking platforms to provide relevant performance metrics, based on different use-cases, that would evolve together with the AMR field; second, agreed and defined datasets to allow the pipelines’ implementation, validation, and quality-control over time. Following previous discussions on the main challenges linked to this approach, we provide concrete recommendations and future steps, related to different aspects of the design of benchmarks, such as the selection and the characteristics of the datasets (quality, choice of pathogens and resistances, etc.), the evaluation criteria of the pipelines, and the way these resources should be deployed in the community.

Strengths and caveats of identifying resistance genes from whole genome sequencing data

INTRODUCTION

Antimicrobial resistance (AMR) continues to present major challenges to modern healthcare. Recent advances in whole-genome sequencing (WGS) have made the rapid molecular characterization of AMR a realistic possibility for diagnostic laboratories; yet major barriers to clinical implementation exist.

AREAS COVERED

We describe and compare short- and long-read sequencing platforms, typical components of bioinformatics pipelines, tools for AMR gene detection and the relative merits of read- or assembly-based approaches. The challenges of characterizing mobile genetic elements from genomic data are outlined, as well as the complexities inherent to the prediction of phenotypic resistance from WGS. Practical obstacles to implementation in diagnostic laboratories, the critical role of quality control and external quality assurance, as well as standardized reporting standards are also discussed. Future directions, such as the application of machine-learning and artificial intelligence algorithms, linked to clinically meaningful outcomes, may offer a new paradigm for the clinical application of AMR prediction.

EXPERT OPINION

AMR prediction from WGS data presents an exciting opportunity to advance our capacity to comprehensively characterize infectious pathogens in a rapid manner, ultimately aiming to improve patient outcomes. Collaborative efforts between clinicians, scientists, regulatory bodies and healthcare administrators will be critical to achieve the full promise of this approach.

The Genetic Relatedness and Antimicrobial Resistance Patterns of Mastitis-Causing Staphylococcus aureus Strains Isolated from New Zealand Dairy Cattle

Staphylococcus aureus is one of the leading causes of bovine mastitis worldwide and is a common indication for use of antimicrobials on dairy farms. This study aims to investigate the association between on-farm antimicrobial usage and the antimicrobial resistance (AMR) profiles of mastitis-causing S. aureus. Whole-genome sequencing was performed on 57 S. aureus isolates derived from cows with either clinical or subclinical mastitis from 17 dairy herds in New Zealand. The genetic relatedness between isolates was examined using the core single nucleotide polymorphism alignment whilst AMR and virulence genes were identified in-silico. The association between gene presence-absence and sequence type (ST), antimicrobial susceptibility and dry cow therapy treatment was investigated using Scoary. Altogether, eight STs were identified with 61.4% (35/57) belonging to ST-1. Furthermore, 14 AMR-associated genes and 76 virulence-associated genes were identified, with little genetic diversity between isolates belonging to the same ST. Several genes including merR1 which is thought to play a role in ciprofloxacin-resistance were found to be significantly overrepresented in isolates sampled from herds using ampicillin/cloxacillin dry cow therapy. Overall, the presence of resistance genes remains low and current antimicrobial usage patterns do not appear to be driving AMR in S. aureus associated with bovine mastitis.

Whole Genome Sequencing Provides an Added Value to the Investigation of Staphylococcal Food Poisoning Outbreaks

Through staphylococcal enterotoxin (SE) production, Staphylococcus aureus is a common cause of food poisoning. Detection of staphylococcal food poisoning (SFP) is mostly performed using immunoassays, which, however, only detect five of 27 SEs described to date. Polymerase chain reactions are, therefore, frequently used in complement to identify a bigger arsenal of SE at the gene level (se) but are labor-intensive. Complete se profiling of isolates from different sources, i.e., food and human cases, is, however, important to provide an indication of their potential link within foodborne outbreak investigation. In addition to complete se gene profiling, relatedness between isolates is determined with more certainty using pulsed-field gel electrophoresis, Staphylococcus protein A gene typing and other methods, but these are shown to lack resolution. We evaluated how whole genome sequencing (WGS) can offer a solution to these shortcomings. By WGS analysis of a selection of S. aureus isolates, including some belonging to a confirmed foodborne outbreak, its added value as the ultimate multiplexing method was demonstrated. In contrast to PCR-based se gene detection for which primers are sometimes shown to be non-specific, WGS enabled complete se gene profiling with high performance, provided that a database containing reference sequences for all se genes was constructed and employed. The custom compiled database and applied parameters were made publicly available in an online user-friendly interface. As an all-in-one approach with high resolution, WGS additionally allowed inferring correct isolate relationships. The different DNA extraction kits that were tested affected neither se gene profiling nor relatedness determination, which is interesting for data sharing during SFP outbreak investigation. Although confirming the production of enterotoxins remains important for SFP investigation, we delivered a proof-of-concept that WGS is a valid alternative and/or complementary tool for outbreak investigation.

Shotgun-metagenomics based prediction of antibiotic resistance and virulence determinants in Staphylococcus aureus from periprosthetic tissue on blood culture bottles

Shotgun-metagenomics may give valuable clinical information beyond the detection of potential pathogen(s). Identification of antimicrobial resistance (AMR), virulence genes and typing directly from clinical samples has been limited due to challenges arising from incomplete genome coverage. We assessed the performance of shotgun-metagenomics on positive blood culture bottles (n = 19) with periprosthetic tissue for typing and prediction of AMR and virulence profiles in Staphylococcus aureus. We used different approaches to determine if sequence data from reads provides more information than from assembled contigs. Only 0.18% of total reads was derived from human DNA. Shotgun-metagenomics results and conventional method results were consistent in detecting S. aureus in all samples. AMR and known periprosthetic joint infection virulence genes were predicted from S. aureus. Mean coverage depth, when predicting AMR genes was 209 ×. Resistance phenotypes could be explained by genes predicted in the sample in most of the cases. The choice of bioinformatic data analysis approach clearly influenced the results, i.e. read-based analysis was more accurate for pathogen identification, while contigs seemed better for AMR profiling. Our study demonstrates high genome coverage and potential for typing and prediction of AMR and virulence profiles in S. aureus from shotgun-metagenomics data.

Genomic Epidemiology and Global Population Structure of Exfoliative Toxin A-Producing Staphylococcus aureus Strains Associated With Staphylococcal Scalded Skin Syndrome

Staphylococci producing exfoliative toxins are the causative agents of staphylococcal scalded skin syndrome (SSSS). Exfoliative toxin A (ETA) is encoded by eta, which is harbored on a temperate bacteriophage ΦETA. A recent increase in the incidence of SSSS in North America has been observed; yet it is largely unknown whether this is the result of host range expansion of ΦETA or migration and emergence of established lineages. Here, we detail an outbreak investigation of SSSS in a neonatal intensive care unit, for which we applied whole-genome sequencing (WGS) and phylogenetic analysis of Staphylococcus aureus isolates collected from cases and screening of healthcare workers. We identified the causative strain as a methicillin-susceptible S. aureus (MSSA) sequence type 582 (ST582) possessing ΦETA. To then elucidate the global distribution of ΦETA among staphylococci, we used a recently developed tool to query extant bacterial WGS data for biosamples containing eta, which yielded 436 genomes collected between 1994 and 2019 from 32 countries. Applying population genomic analysis, we resolved the global distribution of S. aureus with lysogenized ΦETA and assessed antibiotic resistance determinants as well as the diversity of ΦETA. The population is highly structured with eight dominant sequence clusters (SCs) that generally aligned with S. aureus ST clonal complexes. The most prevalent STs included ST109 (24.3%), ST15 (13.1%), ST121 (10.1%), and ST582 (7.1%). Among strains with available data, there was an even distribution of isolates from carriage and disease. Only the SC containing ST121 had significantly more isolates collected from disease (69%, n = 46) than carriage (31%, n = 21). Further, we identified 10.6% (46/436) of strains as methicillin-resistant S. aureus (MRSA) based on the presence of mecA and the SCCmec element. Assessment of ΦETA diversity based on nucleotide identity revealed 27 phylogroups, and prophage gene content further resolved 62 clusters. ΦETA was relatively stable within lineages, yet prophage variation is geographically structured. This suggests that the reported increase in incidence is associated with migration and expansion of existing lineages, not the movement of ΦETA to new genomic backgrounds. This revised global view reveals that ΦETA is diverse and is widely distributed on multiple genomic backgrounds whose distribution varies geographically.

Evaluation of a fully automated bioinformatics tool to predict antibiotic resistance from MRSA genomes

Objectives

The genetic prediction of phenotypic antibiotic resistance based on analysis of WGS data is becoming increasingly feasible, but a major barrier to its introduction into routine use is the lack of fully automated interpretation tools. Here, we report the findings of a large evaluation of the Next Gen Diagnostics (NGD) automated bioinformatics analysis tool to predict the phenotypic resistance of MRSA.

Methods

MRSA-positive patients were identified in a clinical microbiology laboratory in England between January and November 2018. One MRSA isolate per patient together with all blood culture isolates (total n = 778) were sequenced on the Illumina MiniSeq instrument in batches of 21 clinical MRSA isolates and three controls.

Results

The NGD system activated post-sequencing and processed the sequences to determine susceptible/resistant predictions for 11 antibiotics, taking around 11 minutes to analyse 24 isolates sequenced on a single sequencing run. NGD results were compared with phenotypic susceptibility testing performed by the clinical laboratory using the disc diffusion method and EUCAST breakpoints. Following retesting of discrepant results, concordance between phenotypic results and NGD genetic predictions was 99.69%. Further investigation of 22 isolate genomes associated with persistent discrepancies revealed a range of reasons in 12 cases, but no cause could be found for the remainder. Genetic predictions generated by the NGD tool were compared with predictions generated by an independent research-based informatics approach, which demonstrated an overall concordance between the two methods of 99.97%.

Conclusions

We conclude that the NGD system provides rapid and accurate prediction of the antibiotic susceptibility of MRSA.

Applications of Machine Learning to the Problem of Antimicrobial Resistance: an Emerging Model for Translational Research

Antimicrobial resistance (AMR) remains one of the most challenging phenomena of modern medicine. Machine learning (ML) is a subfield of artificial intelligence that focuses on the development of algorithms that learn how to accurately predict outcome variables using large sets of predictor variables that are typically not hand selected and are minimally curated. Models are parameterized using a training data set and then applied to a test data set on which predictive performance is evaluated. The application of ML algorithms to the problem of AMR has garnered increasing interest in the past 5 years due to the exponential growth of experimental and clinical data, heavy investment in computational capacity, improvements in algorithm performance, and increasing urgency for innovative approaches to reducing the burden of disease. Here, we review the current state of research at the intersection of ML and AMR with an emphasis on three domains of work. The first is the prediction of AMR using genomic data. The second is the use of ML to gain insight into the cellular functions disrupted by antibiotics, which forms the basis for understanding mechanisms of action and developing novel anti-infectives. The third focuses on the application of ML for antimicrobial stewardship using data extracted from the electronic health record. Although the use of ML for understanding, diagnosing, treating, and preventing AMR is still in its infancy, the continued growth of data and interest ensures it will become an important tool for future translational research programs.

Antibiotic resistance: Time of synthesis in a post-genomic age

Antibiotic resistance has been highlighted by international organizations, including World Health Organization, World Bank and United Nations, as one of the most relevant global health problems. Classical approaches to study this problem have focused in infected humans, mainly at hospitals. Nevertheless, antibiotic resistance can expand through different ecosystems and geographical allocations, hence constituting a One-Health, Global-Health problem, requiring specific integrative analytic tools. Antibiotic resistance evolution and transmission are multilayer, hierarchically organized processes with several elements (from genes to the whole microbiome) involved. However, their study has been traditionally gene-centric, each element independently studied. The development of robust-economically affordable whole genome sequencing approaches, as well as other -omic techniques as transcriptomics and proteomics, is changing this panorama. These technologies allow the description of a system, either a cell or a microbiome as a whole, overcoming the problems associated with gene-centric approaches. We are currently at the time of combining the information derived from -omic studies to have a more holistic view of the evolution and spread of antibiotic resistance. This synthesis process requires the accurate integration of -omic information into computational models that serve to analyse the causes and the consequences of acquiring AR, fed by curated databases capable of identifying the elements involved in the acquisition of resistance. In this review, we analyse the capacities and drawbacks of the tools that are currently in use for the global analysis of AR, aiming to identify the more useful targets for effective corrective interventions.

Pseudomonas aeruginosa: a clinical and genomics update

Antimicrobial resistance (AMR) has become a global medical priority that needs urgent resolution. Pseudomonas aeruginosa is a versatile, adaptable bacterial species with widespread environmental occurrence, strong medical relevance, a diverse set of virulence genes and a multitude of intrinsic and possibly acquired antibiotic resistance traits. P. aeruginosa causes a wide variety of infections and has an epidemic-clonal population structure. Several of its dominant global clones have collected a wide variety of resistance genes rendering them multi-drug resistant (MDR) and particularly threatening groups of vulnerable individuals including surgical patients, immunocompromised patients, Caucasians suffering from cystic fibrosis (CF) and more. AMR and MDR especially are particularly problematic in P. aeruginosa significantly complicating successful antibiotic treatment. In addition, antimicrobial susceptibility testing (AST) of P. aeruginosa can be cumbersome due to its slow growth or the massive production of exopolysaccharides and other extracellular compounds. For that reason, phenotypic AST is progressively challenged by genotypic methods using whole genome sequences (WGS) and large-scale phenotype databases as a framework of reference. We here summarize the state of affairs and the quality level of WGS-based AST for P. aeruginosa mostly from clinical origin.

Twenty-first century molecular methods for analyzing antimicrobial resistance in surface waters to support One Health assessments

Antimicrobial resistance (AMR) in the environment is a growing global health concern, especially the dissemination of AMR into surface waters due to human and agricultural inputs. Within recent years, research has focused on trying to understand the impact of AMR in surface waters on human, agricultural and ecological health (One Health). While surface water quality assessments and surveillance of AMR have historically utilized culture-based methods, culturing bacteria has limitations due to difficulty in isolating environmental bacteria and the need for a priori information about the bacteria for selective isolation. The use of molecular techniques to analyze AMR at the genetic level has helped to overcome the difficulties with culture-based techniques since they do not require advance knowledge of the bacterial population and can analyze uncultivable environmental bacteria. The aim of this review is to provide an overview of common contemporary molecular methods available for analyzing AMR in surface waters, which include high throughput real-time polymerase chain reaction (HT-qPCR), metagenomics, and whole genome sequencing. This review will also feature how these methods may provide information on human and animal health risks. HT-qPCR works at the nanoliter scale, requires only a small amount of DNA, and can analyze numerous gene targets simultaneously, but may lack in analytical sensitivity and the ability to optimize individual assays compared to conventional qPCR. Metagenomics offers more detailed genomic information and taxonomic resolution than PCR by sequencing all the microbial genomes within a sample. Its open format allows for the discovery of new antibiotic resistance genes; however, the quantity of DNA necessary for this technique can be a limiting factor for surface water samples that typically have low numbers of bacteria per sample volume. Whole genome sequencing provides the complete genomic profile of a single environmental isolate and can identify all genetic elements that may confer AMR. However, a main disadvantage of this technique is that it only provides information about one bacterial isolate and is challenging to utilize for community analysis. While these contemporary techniques can quickly provide a vast array of information about AMR in surface waters, one technique does not fully characterize AMR nor its potential risks to human, animal, or ecological health. Rather, a combination of techniques (including both molecular- and culture-based) are necessary to fully understand AMR in surface waters from a One Health perspective.

Draft Genome Sequences of 62 Staphylococcus aureus Isolates Associated with Four Foodborne Outbreaks in the United States

Staphylococcus aureus bacteria are ranked among the top five foodborne pathogens in the United States. Here, we report the draft genome sequences of 62 S. aureus isolates that originated from the manufacturing environment of an Illinois bakery and were associated with outbreaks between 2010 and 2011 in the United States.

Staphylococcus aureus bacteria are ranked among the top five foodborne pathogens in the United States. Here, we report the draft genome sequences of 62 S. aureus isolates that originated from the manufacturing environment of an Illinois bakery and were associated with outbreaks between 2010 and 2011 in the United States.

Genomic investigation of clinically significant coagulase-negative staphylococci

Introduction. Coagulase-negative staphylococci have been recognized both as emerging pathogens and contaminants of clinical samples. High-resolution genomic investigation may provide insights into their clinical significance.Aims. To review the literature regarding coagulase-negative staphylococcal infection and the utility of genomic methods to aid diagnosis and management, and to identify promising areas for future research.Methodology. We searched Google Scholar with the terms (Staphylococcus) AND (sequencing OR (infection)). We prioritized papers that addressed coagulase-negative staphylococci, genomic analysis, or infection.Results. A number of studies have investigated specimen-related, phenotypic and genetic factors associated with colonization, infection and virulence, but diagnosis remains problematic.Conclusion. Genomic investigation provides insights into the genetic diversity and natural history of colonization and infection. Such information allows the development of new methodologies to identify and compare relatedness and predict antimicrobial resistance. Future clinical studies that employ suitable sampling frames coupled with the application of high-resolution whole-genome sequencing may aid the development of more discriminatory diagnostic approaches to coagulase-staphylococcal infection.

Rapid Methods for Antimicrobial Resistance Diagnostics

Antimicrobial resistance (AMR) is one of the most challenging threats in public health; thus, there is a growing demand for methods and technologies that enable rapid antimicrobial susceptibility testing (AST). The conventional methods and technologies addressing AMR diagnostics and AST employed in clinical microbiology are tedious, with high turnaround times (TAT), and are usually expensive. As a result, empirical antimicrobial therapies are prescribed leading to AMR spread, which in turn causes higher mortality rates and increased healthcare costs. This review describes the developments in current cutting-edge methods and technologies, organized by key enabling research domains, towards fighting the looming AMR menace by employing recent advances in AMR diagnostic tools. First, we summarize the conventional methods addressing AMR detection, surveillance, and AST. Thereafter, we examine more recent non-conventional methods and the advancements in each field, including whole genome sequencing (WGS), matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) spectrometry, Fourier transform infrared (FTIR) spectroscopy, and microfluidics technology. Following, we provide examples of commercially available diagnostic platforms for AST. Finally, perspectives on the implementation of emerging concepts towards developing paradigm-changing technologies and methodologies for AMR diagnostics are discussed.

A roadmap for the generation of benchmarking resources for antimicrobial resistance detection using next generation sequencing

Next Generation Sequencing technologies significantly impact the field of Antimicrobial Resistance (AMR) detection and monitoring, with immediate uses in diagnosis and risk assessment. For this application and in general, considerable challenges remain in demonstrating sufficient trust to act upon the meaningful information produced from raw data, partly because of the reliance on bioinformatics pipelines, which can produce different results and therefore lead to different interpretations. With the constant evolution of the field, it is difficult to identify, harmonise and recommend specific methods for large-scale implementations over time. In this article, we propose to address this challenge through establishing a transparent, performance-based, evaluation approach to provide flexibility in the bioinformatics tools of choice, while demonstrating proficiency in meeting common performance standards. The approach is two-fold: first, a community-driven effort to establish and maintain “live” (dynamic) benchmarking platforms to provide relevant performance metrics, based on different use-cases, that would evolve together with the AMR field; second, agreed and defined datasets to allow the pipelines’ implementation, validation, and quality-control over time. Following previous discussions on the main challenges linked to this approach, we provide concrete recommendations and future steps, related to different aspects of the design of benchmarks, such as the selection and the characteristics of the datasets (quality, choice of pathogens and resistances, etc.), the evaluation criteria of the pipelines, and the way these resources should be deployed in the community.

Whole-genome sequencing as part of national and international surveillance programmes for antimicrobial resistance: a roadmap

The global spread of antimicrobial resistance (AMR) and lack of novel alternative treatments have been declared a global public health emergency by WHO. The greatest impact of AMR is experienced in resource-poor settings, because of lack of access to alternative antibiotics and because the prevalence of multidrug-resistant bacterial strains may be higher in low-income and middle-income countries (LMICs). Intelligent surveillance of AMR infections is key to informed policy decisions and public health interventions to counter AMR. Molecular surveillance using whole-genome sequencing (WGS) can be a valuable addition to phenotypic surveillance of AMR. WGS provides insights into the genetic basis of resistance mechanisms, as well as pathogen evolution and population dynamics at different spatial and temporal scales. Due to its high cost and complexity, WGS is currently mainly carried out in high-income countries. However, given its potential to inform national and international action plans against AMR, establishing WGS as a surveillance tool in LMICs will be important in order to produce a truly global picture. Here, we describe a roadmap for incorporating WGS into existing AMR surveillance frameworks, including WHO Global Antimicrobial Resistance Surveillance System, informed by our ongoing, practical experiences developing WGS surveillance systems in national reference laboratories in Colombia, India, Nigeria and the Philippines. Challenges and barriers to WGS in LMICs will be discussed together with a roadmap to possible solutions.

Concordance of disk diffusion, broth microdilution, and whole-genome sequencing for determination of in vitro antimicrobial susceptibility of Mannheimia haemolytica

Background

Extensive drug resistance (XDR) is an emerging concern with Mannheimia haemolytica, and a variety of testing methods are available for characterizing in vitro antimicrobial susceptibility.

Objectives

To compare the concordance among disk diffusion, broth microdilution, and whole genome sequencing (WGS) for susceptibility testing of M. haemolytica before and after mass treatment using tulathromycin.

Animals

Forty‐eight M. haemolytica isolates collected from high‐risk beef stocker calves before and after mass treatment (metaphylaxis) using tulathromycin (Draxxin, Zoetis, Parsippany, NJ) given at the label dosage of 2.5 mg/kg body weight SC in the neck.

Methods

In vitro antimicrobial susceptibility was determined for all 48 isolates using disk diffusion, broth microdilution, and WGS. Concordance was calculated between pairs of susceptibility testing methods as follows: number of isolates classified identically by the 2 testing methods for each timepoint, divided by the number of isolates tested at that timepoint. Discordance was calculated as follows: number of isolates classified differently by the 2 testing methods for each timepoint, divided by the number of isolates tested at that timepoint.

Results

Concordance between testing methods ranged from 42.3% to 100%, depending on antimicrobial evaluated, timing of sample collection, and testing method used. Very major errors were identified in up to 7.7% of classifications whereas minor errors were seen in up to 50% of classifications depending on antimicrobial evaluated, timing of sample collection, and testing method used.

Conclusions and Clinical Importance

Our results show that discrepancies in the results of different susceptibility testing methods occur and suggest a need for greater harmonization of susceptibility testing methods.

Next-Generation Sequencing in Clinical Microbiology: Are We There Yet?

Next-generation sequencing (NGS) applications have been transitioning from research tools to diagnostic methods and are becoming more commonplace in clinical microbiology laboratories. These applications include (1) whole-genome sequencing, (2) targeted next-generation sequencing methods, and (3) metagenomic next-generation sequencing. The introduction of these methods into the clinical microbiology laboratory has led to the theoretic question of "Will NGS-based methods supplant traditional methods for strain typing, identification, and antimicrobial susceptibility prediction?" The authors address this question and discuss where we are at now with clinical NGS applications for infectious diseases, what does the future hold, and at what cost?

NAuRA: Genomic Tool to Identify Staphylococcal Enterotoxins in Staphylococcus aureus Strains Responsible for FoodBorne Outbreaks

Food contamination by staphylococcal enterotoxins (SEs) is responsible for many food poisoning outbreaks (FPOs) each year, and they represent the third leading cause of FPOs in Europe. SEs constitute a protein family with 27 proteins. However, enzyme immunoassays can only detect directly in food the five classical SEs (SEA-SEE). Thus, molecular characterization methods of strains found in food are now used for FPO investigations. Here, we describe the development and implementation of a genomic analysis tool called NAuRA (Nice automatic Research of alleles) that can detect the presence of 27 SEs genes in just one analysis- and create a database of allelic data and protein variants for harmonizing analyses. This tool uses genome assembly data and the 27 protein sequences of SEs. To include the different divergence levels between SE-coding genes, parameters of coverage and identity were generated from 10,000 simulations and a dataset of 244 assembled genomes from strains responsible for outbreaks in Europe as well as the RefSeq reference database. Based on phylogenetic inference performed using maximum-likelihood on the core genomes of the strains in this collection, we demonstrated that strains responsible for FPOs are distributed throughout the phylogenetic tree. Moreover, 71 toxin profiles were obtained using the NAuRA pipeline and these profiles do not follow the evolutionary history of strains. This study presents a pioneering method to investigate strains isolated from food at the genomic level and to analyze the diversity of all 27 SE-coding genes together.

Platforms for elucidating antibiotic resistance in single genomes and complex metagenomes

Antibiotic or antimicrobial resistance (AR) facilitated by the vertical and/or horizontal transfer of antibiotic resistance genes (ARGs), is a serious global health challenge. While traditionally associated with pathogens in clinical environments, it is becoming increasingly clear that non-clinical environments may also be reservoirs of ARGs. The recent improvements in rapid and affordable next generation sequencing technologies along with sophisticated bioinformatics platforms has the potential to revolutionize diagnostic microbiology and microbial surveillance. Through the study and characterization of ARGs in bacterial genomes and complex metagenomes, we are now able to reveal the genetic scope of AR in single bacteria and complex communities, and obtain important insights into AR dynamics at species, population and community levels, providing novel epidemiological and ecological perspectives. A suite of bioinformatics pipelines and ARG databases are currently available for genomic and metagenomic data analyses. However, different platforms may significantly vary and therefore, it is crucial to choose the tools that are most suitable for the specific analysis being conducted. This review provides a detailed account of available bioinformatics platforms for identification and characterization of ARGs and associated genetic elements within single bacterial isolates and complex environmental samples. It focuses primarily on currently available ARG databases, employing a comprehensive benchmarking pipeline to identify ARGs in four bacterial genomes (Aeromonas salmonicida, Bacillus cereus, Burkholderia sp. and Escherichia coli) and three shotgun metagenomes (human gut, poultry litter and soil) providing insight into which databases should be used for different analytical scenarios.

Staphylococcus aureus whole genome sequence-based susceptibility and resistance prediction using a clinically amenable workflow

We used graphical user interface-based automated analytical tools from Next Gen Diagnostics (Mountain View, CA) and 1928 Diagnostics (Gothenburg, Sweden) to analyze whole genome sequence (WGS) data from 102 unique blood culture isolates of Staphylococcus aureus to predict antimicrobial susceptibly, with results compared to those of phenotypic susceptibility testing. Of 916 isolate/antibiotic combinations analyzed using the Next Gen Diagnostics tool, there were 9 discrepancies between WGS predictions and phenotypic susceptibility/resistance, including 8 for clindamycin and 1 for minocycline. Of 612 isolate/antibiotic combinations analyzed using the 1928 Diagnostics tool, there were 13 discrepancies between WGS predictions and phenotypic susceptibility/resistance, including 9 for clindamycin, 3 for trimethoprim-sulfamethoxazole, and 1 for rifampin. Trimethoprim-sulfamethoxazole was not assessed by Next Gen Diagnostics, and minocycline was not assessed by 1928 Diagnostics. There was complete concordance between phenotypic susceptibility/resistance and genotypic prediction of susceptibility/resistance using both analytical platforms for oxacillin, vancomycin, and mupirocin, as well as by the Next Gen Diagnostics analytical tool for levofloxacin (the 1928 Diagnostics tool did not assess levofloxacin). These results suggest that, from a performance standpoint, with some caveats, automatic bioinformatics tools may be acceptable to predict susceptibility and resistance to a panel of antibiotics for S. aureus.

Innovative and rapid antimicrobial susceptibility testing systems

Antimicrobial resistance (AMR) is a major threat to human health worldwide, and the rapid detection and quantification of resistance, combined with antimicrobial stewardship, are key interventions to combat the spread and emergence of AMR. Antimicrobial susceptibility testing (AST) systems are the collective set of diagnostic processes that facilitate the phenotypic and genotypic assessment of AMR and antibiotic susceptibility. Over the past 30 years, only a few high-throughput AST methods have been developed and widely implemented. By contrast, several studies have established proof of principle for various innovative AST methods, including both molecular-based and genome-based methods, which await clinical trials and regulatory review. In this Review, we discuss the current state of AST systems in the broadest technical, translational and implementation-related scope.

Discordant bioinformatic predictions of antimicrobial resistance from whole-genome sequencing data of bacterial isolates: an inter-laboratory study

Antimicrobial resistance (AMR) poses a threat to public health. Clinical microbiology laboratories typically rely on culturing bacteria for antimicrobial-susceptibility testing (AST). As the implementation costs and technical barriers fall, whole-genome sequencing (WGS) has emerged as a 'one-stop' test for epidemiological and predictive AST results. Few published comparisons exist for the myriad analytical pipelines used for predicting AMR. To address this, we performed an inter-laboratory study providing sets of participating researchers with identical short-read WGS data from clinical isolates, allowing us to assess the reproducibility of the bioinformatic prediction of AMR between participants, and identify problem cases and factors that lead to discordant results. We produced ten WGS datasets of varying quality from cultured carbapenem-resistant organisms obtained from clinical samples sequenced on either an Illumina NextSeq or HiSeq instrument. Nine participating teams ('participants') were provided these sequence data without any other contextual information. Each participant used their choice of pipeline to determine the species, the presence of resistance-associated genes, and to predict susceptibility or resistance to amikacin, gentamicin, ciprofloxacin and cefotaxime. We found participants predicted different numbers of AMR-associated genes and different gene variants from the same clinical samples. The quality of the sequence data, choice of bioinformatic pipeline and interpretation of the results all contributed to discordance between participants. Although much of the inaccurate gene variant annotation did not affect genotypic resistance predictions, we observed low specificity when compared to phenotypic AST results, but this improved in samples with higher read depths. Had the results been used to predict AST and guide treatment, a different antibiotic would have been recommended for each isolate by at least one participant. These challenges, at the final analytical stage of using WGS to predict AMR, suggest the need for refinements when using this technology in clinical settings. Comprehensive public resistance sequence databases, full recommendations on sequence data quality and standardization in the comparisons between genotype and resistance phenotypes will all play a fundamental role in the successful implementation of AST prediction using WGS in clinical microbiology laboratories.

Detailed Evaluation of Data Analysis Tools for Subtyping of Bacterial Isolates Based on Whole Genome Sequencing: Neisseria meningitidis as a Proof of Concept

Whole genome sequencing is increasingly recognized as the most informative approach for characterization of bacterial isolates. Success of the routine use of this technology in public health laboratories depends on the availability of well-characterized and verified data analysis methods. However, multiple subtyping workflows are now often being used for a single organism, and differences between them are not always well described. Moreover, methodologies for comparison of subtyping workflows, and assessment of their performance are only beginning to emerge. Current work focuses on the detailed comparison of WGS-based subtyping workflows and evaluation of their suitability for the organism and the research context in question. We evaluated the performance of pipelines used for subtyping of Neisseria meningitidis, including the currently widely applied cgMLST approach and different SNP-based methods. In addition, the impact of the use of different tools for detection and filtering of recombinant regions and of different reference genomes were tested. Our benchmarking analysis included both assessment of technical performance of the pipelines and functional comparison of the generated genetic distance matrices and phylogenetic trees. It was carried out using replicate sequencing datasets of high- and low-coverage, consisting mainly of isolates belonging to the clonal complex 269. We demonstrated that cgMLST and some of the SNP-based subtyping workflows showed very good performance characteristics and highly similar genetic distance matrices and phylogenetic trees with isolates belonging to the same clonal complex. However, only two of the tested workflows demonstrated reproducible results for a group of more closely related isolates. Additionally, results of the SNP-based subtyping workflows were to some level dependent on the reference genome used. Interestingly, the use of recombination-filtering software generally reduced the similarity between the gene-by-gene and SNP-based methodologies for subtyping of N. meningitidis. Our study, where N. meningitidis was taken as an example, clearly highlights the need for more benchmarking comparative studies to eventually contribute to a justified use of a specific WGS data analysis workflow within an international public health laboratory context.

Whole genome sequencing and metagenomics for outbreak investigation, source attribution and risk assessment of food-borne microorganisms

This Opinion considers the application of whole genome sequencing (WGS) and metagenomics for outbreak investigation, source attribution and risk assessment of food‐borne pathogens. WGS offers the highest level of bacterial strain discrimination for food‐borne outbreak investigation and source‐attribution as well as potential for more precise hazard identification, thereby facilitating more targeted risk assessment and risk management. WGS improves linking of sporadic cases associated with different food products and geographical regions to a point source outbreak and can facilitate epidemiological investigations, allowing also the use of previously sequenced genomes. Source attribution may be favoured by improved identification of transmission pathways, through the integration of spatial‐temporal factors and the detection of multidirectional transmission and pathogen–host interactions. Metagenomics has potential, especially in relation to the detection and characterisation of non‐culturable, difficult‐to‐culture or slow‐growing microorganisms, for tracking of hazard‐related genetic determinants and the dynamic evaluation of the composition and functionality of complex microbial communities. A SWOT analysis is provided on the use of WGS and metagenomics for Salmonella and Shigatoxin‐producing Escherichia coli (STEC) serotyping and the identification of antimicrobial resistance determinants in bacteria. Close agreement between phenotypic and WGS‐based genotyping data has been observed. WGS provides additional information on the nature and localisation of antimicrobial resistance determinants and on their dissemination potential by horizontal gene transfer, as well as on genes relating to virulence and biological fitness. Interoperable data will play a major role in the future use of WGS and metagenomic data. Capacity building based on harmonised, quality controlled operational systems within European laboratories and worldwide is essential for the investigation of cross‐border outbreaks and for the development of international standardised risk assessments of food‐borne microorganisms.

Antibiotic resistance prediction for Mycobacterium tuberculosis from genome sequence data with Mykrobe

Two billion people are infected with Mycobacterium tuberculosis, leading to 10 million new cases of active tuberculosis and 1.5 million deaths annually. Universal access to drug susceptibility testing (DST) has become a World Health Organization priority. We previously developed a software tool, Mykrobe predictor, which provided offline species identification and drug resistance predictions for M. tuberculosis from whole genome sequencing (WGS) data. Performance was insufficient to support the use of WGS as an alternative to conventional phenotype-based DST, due to mutation catalogue limitations.  Here we present a new tool, Mykrobe, which provides the same functionality based on a new software implementation. Improvements include i) an updated mutation catalogue giving greater sensitivity to detect pyrazinamide resistance, ii) support for user-defined resistance catalogues, iii) improved identification of non-tuberculous mycobacterial species, and iv) an updated statistical model for Oxford Nanopore Technologies sequencing data. Mykrobe is released under MIT license at https://github.com/mykrobe-tools/mykrobe. We incorporate mutation catalogues from the CRyPTIC consortium et al. (2018) and from Walker et al. (2015), and make improvements based on performance on an initial set of 3206 and an independent set of 5845 M. tuberculosis Illumina sequences. To give estimates of error rates, we use a prospectively collected dataset of 4362 M. tuberculosis isolates. Using culture based DST as the reference, we estimate Mykrobe to be 100%, 95%, 82%, 99% sensitive and 99%, 100%, 99%, 99% specific for rifampicin, isoniazid, pyrazinamide and ethambutol resistance prediction respectively. We benchmark against four other tools on 10207 (=5845+4362) samples, and also show that Mykrobe gives concordant results with nanopore data.  We measure the ability of Mykrobe-based DST to guide personalized therapeutic regimen design in the context of complex drug susceptibility profiles, showing 94% concordance of implied regimen with that driven by phenotypic DST, higher than all other benchmarked tools.

Using Genomics to Track Global Antimicrobial Resistance

The recent advancements in rapid and affordable DNA sequencing technologies have revolutionized diagnostic microbiology and microbial surveillance. The availability of bioinformatics tools and online accessible databases has been a prerequisite for this. We conducted a scientific literature review and here we present a description of examples of available tools and databases for antimicrobial resistance (AMR) detection and provide future perspectives and recommendations. At least 47 freely accessible bioinformatics resources for detection of AMR determinants in DNA or amino acid sequence data have been developed to date. These include, among others but not limited to, ARG-ANNOT, CARD, SRST2, MEGARes, Genefinder, ARIBA, KmerResistance, AMRFinder, and ResFinder. Bioinformatics resources differ for several parameters including type of accepted input data, presence/absence of software for search within a database of AMR determinants that can be specific to a tool or cloned from other resources, and for the search approach employed, which can be based on mapping or on alignment. As a consequence, each tool has strengths and limitations in sensitivity and specificity of detection of AMR determinants and in application, which for some of the tools have been highlighted in benchmarking exercises and scientific articles. The identified tools are either available at public genome data centers, from GitHub or can be run locally. NCBI and European Nucleotide Archive (ENA) provide possibilities for online submission of both sequencing and accompanying phenotypic antimicrobial susceptibility data, allowing for other researchers to further analyze data, and develop and test new tools. The advancement in whole genome sequencing and the application of online tools for real-time detection of AMR determinants are essential to identify control and prevention strategies to combat the increasing threat of AMR. Accessible tools and DNA sequence data are expanding, which will allow establishing global pathogen surveillance and AMR tracking based on genomics. There is however, a need for standardization of pipelines and databases as well as phenotypic predictions based on the data.

Evaluation of parameters affecting performance and reliability of machine learning-based antibiotic susceptibility testing from whole genome sequencing data

Prediction of antibiotic resistance phenotypes from whole genome sequencing data by machine learning methods has been proposed as a promising platform for the development of sequence-based diagnostics. However, there has been no systematic evaluation of factors that may influence performance of such models, how they might apply to and vary across clinical populations, and what the implications might be in the clinical setting. Here, we performed a meta-analysis of seven large Neisseria gonorrhoeae datasets, as well as Klebsiella pneumoniae and Acinetobacter baumannii datasets, with whole genome sequence data and antibiotic susceptibility phenotypes using set covering machine classification, random forest classification, and random forest regression models to predict resistance phenotypes from genotype. We demonstrate how model performance varies by drug, dataset, resistance metric, and species, reflecting the complexities of generating clinically relevant conclusions from machine learning-derived models. Our findings underscore the importance of incorporating relevant biological and epidemiological knowledge into model design and assessment and suggest that doing so can inform tailored modeling for individual drugs, pathogens, and clinical populations. We further suggest that continued comprehensive sampling and incorporation of up-to-date whole genome sequence data, resistance phenotypes, and treatment outcome data into model training will be crucial to the clinical utility and sustainability of machine learning-based molecular diagnostics.

A large scale evaluation of TBProfiler and Mykrobe for antibiotic resistance prediction in Mycobacterium tuberculosis

Recent years saw a growing interest in predicting antibiotic resistance from whole-genome sequencing data, with promising results obtained for Staphylococcus aureus and Mycobacterium tuberculosis. In this work, we gathered 6,574 sequencing read datasets of M. tuberculosis public genomes with associated antibiotic resistance profiles for both first and second-line antibiotics. We performed a systematic evaluation of TBProfiler and Mykrobe, two widely recognized softwares allowing to predict resistance in M. tuberculosis. The size of the dataset allowed us to obtain confident estimations of their overall predictive performance, to assess precisely the individual predictive power of the markers they rely on, and to study in addition how these softwares behave across the major M. tuberculosis lineages. While this study confirmed the overall good performance of these tools, it revealed that an important fraction of the catalog of mutations they embed is of limited predictive power. It also revealed that these tools offer different sensitivity/specificity trade-offs, which is mainly due to the different sets of mutation they embed but also to their underlying genotyping pipelines. More importantly, it showed that their level of predictive performance varies greatly across lineages for some antibiotics, therefore suggesting that the predictions made by these softwares should be deemed more or less confident depending on the lineage inferred and the predictive performance of the marker(s) actually detected. Finally, we evaluated the relevance of machine learning approaches operating from the set of markers detected by these softwares and show that they present an attractive alternative strategy, allowing to reach better performance for several drugs while significantly reducing the number of candidate mutations to consider.

Review of molecular subtyping methodologies used to investigate outbreaks due to multidrug-resistant enteric bacterial pathogens in sub-Saharan Africa

Background

In sub-Saharan Africa, molecular epidemiological investigation of outbreaks caused by antimicrobial-resistant enteric bacterial pathogens have mostly been described for Salmonella species, Vibrio cholerae, Shigella species and Escherichia coli. For these organisms, I reviewed all publications describing the use of molecular subtyping methodologies to investigate outbreaks caused by multidrug-resistant (MDR) enteric bacterial infections.

Objectives

To describe the use of molecular subtyping methodologies to investigate outbreaks caused by MDR enteric bacterial pathogens in sub-Saharan Africa and to describe the current status of molecular subtyping capabilities in the region.

Methods

A PubMed database literature search (English language only) was performed using the search strings: ‘Africa outbreak MDR’, ‘Africa outbreak multi’, ‘Africa outbreak multidrug’, ‘Africa outbreak multi drug’, ‘Africa outbreak resistance’, ‘Africa outbreak resistant’, ‘Africa outbreak drug’, ‘Africa outbreak antibiotic’, ‘Africa outbreak antimicrobial’. These search strings were used in combination with genus and species names of the organisms listed above. All results were included in the review.

Results

The year 1991 saw one of the first reports describing the use of molecular subtyping methodologies in sub-Saharan Africa; this included the use of plasmid profiling to characterise Salmonella Enteritidis. To date, several methodologies have been used; pulsed-field gel electrophoresis analysis and multilocus sequence typing have been the most commonly used methodologies. Investigations have particularly highlighted the emergence and spread of MDR clones; these include Salmonella Typhi H58 and Salmonella Typhimurium ST313 clones. In recent times, whole-genome sequencing (WGS) analysis approaches have increasingly been used.

Conclusion

Traditional molecular subtyping methodologies are still commonly used and still have their place in investigations; however, WGS approaches have increasingly been used and are slowly gaining a stronghold. African laboratories need to start adapting their molecular surveillance methodologies to include WGS, as it is foreseen that WGS analysis will eventually replace all traditional methodologies.

Genome-Based Prediction of Bacterial Antibiotic Resistance

Clinical microbiology has long relied on growing bacteria in culture to determine antimicrobial susceptibility profiles, but the use of whole-genome sequencing for antibiotic susceptibility testing (WGS-AST) is now a powerful alternative. This review discusses the technologies that made this possible and presents results from recent studies to predict resistance based on genome sequences. We examine differences between calling antibiotic resistance profiles by the simple presence or absence of previously known genes and single-nucleotide polymorphisms (SNPs) against approaches that deploy machine learning and statistical models. Often, the limitations to genome-based prediction arise from limitations of accuracy of culture-based AST in addition to an incomplete knowledge of the genetic basis of resistance. However, we need to maintain phenotypic testing even as genome-based prediction becomes more widespread to ensure that the results do not diverge over time. We argue that standardization of WGS-AST by challenge with consistently phenotyped strain sets of defined genetic diversity is necessary to compare the efficacy of methods of prediction of antibiotic resistance based on genome sequences.

Lowering the Barriers to Routine Whole-Genome Sequencing of Bacteria in the Clinical Microbiology Laboratory

Whole-genome sequencing of bacterial isolates is increasingly being used to predict antibacterial susceptibility and resistance. Mason and coauthors describe the phenotypic susceptibility interpretations of more than 1,300 Staphylococcus aureus isolates tested against a dozen antistaphylococcal agents, and they compared these findings to susceptibility predictions made by analyzing whole-genome sequence data (J Clin Microbiol 56:e01815-17, 2018, https://doi.org/10.1128/JCM.01815-17). The genotype-phenotype susceptibility interpretations correlated in 96.3% (2,720/2,825) of resistant findings and 98.8% (11,504/11,639) of susceptible findings. This work by Mason and colleagues is helping to lower the barriers to using whole-genome sequencing of S. aureus in clinical microbiology practice.