StrainSeeker: fast identification of bacterial strains from raw sequencing reads using user-provided guide trees

A survey of k-mer methods and applications in bioinformatics

The rapid progression of genomics and proteomics has been driven by the advent of advanced sequencing technologies, large, diverse, and readily available omics datasets, and the evolution of computational data processing capabilities. The vast amount of data generated by these advancements necessitates efficient algorithms to extract meaningful information. K-mers serve as a valuable tool when working with large sequencing datasets, offering several advantages in computational speed and memory efficiency and carrying the potential for intrinsic biological functionality. This review provides an overview of the methods, applications, and significance of k-mers in genomic and proteomic data analyses, as well as the utility of absent sequences, including nullomers and nullpeptides, in disease detection, vaccine development, therapeutics, and forensic science. Therefore, the review highlights the pivotal role of k-mers in addressing current genomic and proteomic problems and underscores their potential for future breakthroughs in research.

Disseminated Infection Caused by Nocardia cyriacigeorgica in Immunocompromised Patient Confirmed by Whole Genome Sequencing

INTRODUCTION

Nocardia spp. is an opportunistic pathogen capable of causing localized and disseminated infections in immunocompromised hosts. It is critical for serious infections to have an early and accurate identification of this pathogen in order to enable timely and focused combination antimicrobial treatment.

CASE PRESENTATION

We describe the case of an 87-year-old patient previously treated for myasthenia gravis with corticosteroids and azathioprine. Patient was admitted at the emergency department with clinical signs of sepsis with cellulitis of right hand associated with injury acquired after gardening and trimming roses and did not respond to empirical antimicrobial treatment. Computerized tomography revealed pulmonary infiltrates with inflammatory etiology. Nocardia cyriacigeorgica was cultivated from blood culture, skin swab, abscess aspirate, and endotracheal aspirate and identified by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS), 16S rRNA sequencing, and whole genome sequencing (WGS). Susceptibility testing was performed with E-test (bioMerieux, Marcy-l'Étoile, France), and corresponding resistance genes were detected by WGS. Resistance to amoxicillin-clavulanate, azithromycin, ciprofloxacin, and vancomycin was detected by both methods. Despite all interventions and the patient receiving antimicrobial treatment including imipenem-cilastatin, amikacin, and trimethoprim-sulfamethoxazole, the course and outcome of infection were unfavorable.

CONCLUSION

We would like to emphasize the need to consider the possibility of disseminated Nocardia infection in immunocompromised patients, especially in patients receiving long-term corticosteroid treatment with skin infections and/or cavitary lung lesions, especially if these do not improve with standard antimicrobial treatment. Precise species identity provides a critical guide for physicians in the choice of targeted treatment. Thanks to MALDI-TOF MS, Nocardia spp. identification is now available in routine lab work. WGS is still inevitable for the identification of uncommon and novel species due to the high sequence similarities between closely related species and the genetic diversity of that genus.

Increasing Incidence of Streptococcus anginosus Group Intracranial Infections Associated With Sinusitis, Otitis Media, and Mastoiditis in Children

BACKGROUND

The Streptococcus anginosus group (SAG) pathogens have the potential to cause head and neck space infections, including intracranial abscesses. Several centers noted an increase in intracranial abscesses in children during the SARS-CoV-2 pandemic, prompting a Centers for Disease Control and Prevention health alert in May 2022. We examined the epidemiology of pediatric intracranial abscesses at a tertiary care center with a focus on SAG pre- and post-pandemic.

METHODS

Cases of intracranial abscesses of any microbiologic etiology admitted from January 2011 to December 2022 were identified using International Classification of Diseases 10 codes. Subjects were cross-referenced with culture results from the microbiology laboratory at Texas Children's Hospital. Cases included were those associated with either otitis media, mastoiditis or sinusitis and medical records were reviewed.

RESULTS

A total of 157 cases were identified and 59.9% (n = 94) were caused by SAG. The incidence of all sinogenic/otogenic intracranial infections ( P = 0.002), and SAG-specific infections ( P = 0.004), increased from 2011 to 2022. SAG infection was more often associated with multiple surgeries, and these subjects were more likely to require craniotomy or craniectomy. Among sinogenic abscesses, S. intermedius was the most common pathogen, while among otogenic cases, S. pyogenes predominated. From March 2020 to Dec 2022, 9/49 cases tested positive for SARS-CoV-2 (18.4%); characteristics of infection were not significantly different among cases with and without SARS-CoV-2.

CONCLUSIONS

Over the last decade, intracranial complications of sinusitis/otitis have been increasing, specifically those caused by SAG; this trend, however, predated the SARS-CoV-2 pandemic. SAG was associated with a greater need for surgical intervention, specifically neurosurgery. Further work is necessary to determine the cause for these rising infections.

Are the predicted known bacterial strains in a sample really present? A case study

With mutations constantly accumulating in bacterial genomes, it is unclear whether the previously identified bacterial strains are really present in an extant sample. To address this question, we did a case study on the known strains of the bacterial species S. aureus and S. epidermis in 68 atopic dermatitis shotgun metagenomic samples. We evaluated the likelihood of the presence of all sixteen known strains predicted in the original study and by two popular tools in this study. We found that even with the same tool, only two known strains were predicted by the original study and this study. Moreover, none of the sixteen known strains was likely present in these 68 samples. Our study thus indicates the limitation of the known-strain-based studies, especially those on rapidly evolving bacterial species. It implies the unlikely presence of the previously identified known strains in a current environmental sample. It also called for de novo bacterial strain identification directly from shotgun metagenomic reads.

Mixed infections in genotypic drug-resistant Mycobacterium tuberculosis

Tuberculosis disease (TB), caused by Mycobacterium tuberculosis, is a major global public health problem, resulting in more than 1 million deaths each year. Drug resistance (DR), including multi-drug (MDR-TB), is making TB control difficult and accounts for 16% of new and 48% of previously treated cases. To further complicate treatment decision-making, many clinical studies have reported patients harbouring multiple distinct strains of M. tuberculosis across the main lineages (L1 to L4). The extent to which drug-resistant strains can be deconvoluted within mixed strain infection samples is understudied. Here, we analysed M. tuberculosis isolates with whole genome sequencing data (n = 50,723), which covered the main lineages (L1 9.1%, L2 27.6%, L3 11.8%, L4 48.3%), with genotypic resistance to isoniazid (HR-TB; n = 9546 (29.2%)), rifampicin (RR-TB; n = 7974 (24.4%)), and at least MDR-TB (n = 5385 (16.5%)). TB-Profiler software revealed 531 (1.0%) isolates with potential mixed sub-lineage infections, including some with DR mutations (RR-TB 21/531; HR-TB 59/531; at least MDR-TB 173/531). To assist with the deconvolution of such mixtures, we adopted and evaluated a statistical Gaussian Mixture model (GMM) approach. By simulating 240 artificial mixtures of different ratios from empirical data across L1 to L4, a GMM approach was able to accurately estimate the DR profile of each lineage, with a low error rate for the estimated mixing proportions (mean squared error 0.012) and high accuracy for the DR predictions (93.5%). Application of the GMM model to the clinical mixtures (n = 531), found that 33.3% (188/531) of samples consisted of DR and sensitive lineages, 20.2% (114/531) consisted of lineages with only DR mutations, and 40.6% (229/531) consisted of lineages with genotypic pan-susceptibility. Overall, our work demonstrates the utility of combined whole genome sequencing data and GMM statistical analysis approaches for providing insights into mono and mixed M. tuberculosis infections, thereby potentially assisting diagnosis, treatment decision-making, drug resistance and transmission mapping for infection control.

Development of a novel real-time PCR multiplex assay for detection of Streptococcus equi subspecies equi and Streptococcus equi subspecies zooepidemicus

Strangles is a contagious bacterial disease of horses caused by Streptococcus equi subspecies equi (SEE) that occurs globally. Rapid and accurate identification of infected horses is essential for controlling strangles. Because of limitations of existing PCR assays for SEE, we sought to identify novel primers and probes that enable simultaneous detection and differentiation of infection with SEE and S. equi subsp. zooepidemicus (SEZ). Comparative genomics of U.S. strains of SEE and SEZ (n = 50 each) identified SE00768 from SEE and comB from SEZ as target genes. Primers and probes for real-time PCR (rtPCR) were designed for these genes and then aligned in silico with the genomes of strains of SEE (n = 725) and SEZ (n = 343). Additionally, the sensitivity and specificity relative to microbiologic culture were compared between 85 samples submitted to an accredited veterinary medical diagnostic laboratory. The respective primer and probe sets aligned with 99.7 % (723/725) isolates of SEE and 97.1 % (333/343) of SEZ. Of 85 diagnostic samples, 20 of 21 (95.2 %) SEE and 22 of 23 SEZ (95.6 %) culture-positive samples were positive by rtPCR for SEE and SEZ, respectively. Both SEE (n = 2) and SEZ (n = 3) were identified by rtPCR among 32 culture-negative samples. Results were rtPCR-positive for both SEE and SEZ in 21 of 44 (47.7 %) samples that were culture-positive for SEE or SEZ. The primers and probe sets reported here reliably detect SEE and SEZ from Europe and the U.S., and permit detection of concurrent infection with both subspecies.

High-resolution strain-level microbiome composition analysis from short reads

BACKGROUND

Bacterial strains under the same species can exhibit different biological properties, making strain-level composition analysis an important step in understanding the dynamics of microbial communities. Metagenomic sequencing has become the major means for probing the microbial composition in host-associated or environmental samples. Although there are a plethora of composition analysis tools, they are not optimized to address the challenges in strain-level analysis: highly similar strain genomes and the presence of multiple strains under one species in a sample. Thus, this work aims to provide a high-resolution and more accurate strain-level analysis tool for short reads.

RESULTS

In this work, we present a new strain-level composition analysis tool named StrainScan that employs a novel tree-based k-mers indexing structure to strike a balance between the strain identification accuracy and the computational complexity. We tested StrainScan extensively on a large number of simulated and real sequencing data and benchmarked StrainScan with popular strain-level analysis tools including Krakenuniq, StrainSeeker, Pathoscope2, Sigma, StrainGE, and StrainEst. The results show that StrainScan has higher accuracy and resolution than the state-of-the-art tools on strain-level composition analysis. It improves the F1 score by 20% in identifying multiple strains at the strain level.

CONCLUSIONS

By using a novel k-mer indexing structure, StrainScan is able to provide strain-level analysis with higher resolution than existing tools, enabling it to return more informative strain composition analysis in one sample or across multiple samples. StrainScan takes short reads and a set of reference strains as input and its source codes are freely available at https://github.com/liaoherui/StrainScan . Video Abstract.

Transmission of Carbapenem-Resistant Klebsiella pneumoniae in US Hospitals

BACKGROUND

Carbapenem-resistant Klebsiella pneumoniae (CRKp) is the most prevalent carbapenem-resistant Enterobacterales in the United States. We evaluated CRKp clustering in patients in US hospitals.

METHODS

From April 2016 to August 2017, 350 patients with clonal group 258 CRKp were enrolled in the Consortium on Resistance Against Carbapenems in Klebsiella and other Enterobacteriaceae, a prospective, multicenter, cohort study. A maximum likelihood tree was constructed using RAxML. Static clusters shared ≤21 single-nucleotide polymorphisms (SNP) and a most recent common ancestor. Dynamic clusters incorporated SNP distance, culture timing, and rates of SNP accumulation and transmission using the R program TransCluster.

RESULTS

Most patients were admitted from home (n = 150, 43%) or long-term care facilities (n = 115, 33%). Urine (n = 149, 43%) was the most common isolation site. Overall, 55 static and 47 dynamics clusters were identified involving 210 of 350 (60%) and 194 of 350 (55%) patients, respectively. Approximately half of static clusters were identical to dynamic clusters. Static clusters consisted of 33 (60%) intrasystem and 22 (40%) intersystem clusters. Dynamic clusters consisted of 32 (68%) intrasystem and 15 (32%) intersystem clusters and had fewer SNP differences than static clusters (8 vs 9; P = .045; 95% confidence interval [CI]: -4 to 0). Dynamic intersystem clusters contained more patients than dynamic intrasystem clusters (median [interquartile range], 4 [2, 7] vs 2 [2, 2]; P = .007; 95% CI: -3 to 0).

CONCLUSIONS

Widespread intrasystem and intersystem transmission of CRKp was identified in hospitalized US patients. Use of different methods for assessing genetic similarity resulted in only minor differences in interpretation.

Genomic, morphological, and biochemical analyses of a multi-metal resistant but multi-drug susceptible strain of Bordetella petrii from hospital soil

Contamination of soil by antibiotics and heavy metals originating from hospital facilities has emerged as a major cause for the development of resistant microbes. We collected soil samples surrounding a hospital effluent and measured the resistance of bacterial isolates against multiple antibiotics and heavy metals. One strain BMCSI 3 was found to be sensitive to all tested antibiotics. However, it was resistant to many heavy metals and metalloids like cadmium, chromium, copper, mercury, arsenic, and others. This strain was motile and potentially spore-forming. Whole-genome shotgun assembly of BMCSI 3 produced 4.95 Mb genome with 4,638 protein-coding genes. The taxonomic and phylogenetic analysis revealed it, to be a Bordetella petrii strain. Multiple genomic islands carrying mobile genetic elements; coding for heavy metal resistant genes, response regulators or transcription factors, transporters, and multi-drug efflux pumps were identified from the genome. A comparative genomic analysis of BMCSI 3 with annotated genomes of other free-living B. petrii revealed the presence of multiple transposable elements and several genes involved in stress response and metabolism. This study provides insights into how genomic reorganization and plasticity results in evolution of heavy metals resistance by acquiring genes from its natural environment.

Computational analyses of bacterial strains from shotgun reads

Shotgun sequencing is routinely employed to study bacteria in microbial communities. With the vast amount of shotgun sequencing reads generated in a metagenomic project, it is crucial to determine the microbial composition at the strain level. This study investigated 20 computational tools that attempt to infer bacterial strain genomes from shotgun reads. For the first time, we discussed the methodology behind these tools. We also systematically evaluated six novel-strain-targeting tools on the same datasets and found that BHap, mixtureS and StrainFinder performed better than other tools. Because the performance of the best tools is still suboptimal, we discussed future directions that may address the limitations.

Identification of Antimicrobial Compounds in Two Streptomyces sp. Strains Isolated From Beehives

The World Health Organization warns that the alarming increase in antibiotic resistant bacteria will lead to 2.7 million deaths annually due to the lack of effective antibiotic therapies. Clearly, there is an urgent need for short-term alternatives that help to alleviate these alarming figures. In this respect, the scientific community is exploring neglected ecological niches from which the prototypical antibiotic-producing bacteria Streptomycetes are expected to be present. Recent studies have reported that honeybees and their products carry Streptomyces species that possess strong antibacterial activity. In this study, we have investigated the antibiotic profile of two Streptomycetes strains that were isolated from beehives. One of the isolates is the strain Streptomyces albus AN1, which derives from pollen, and shows potent antimicrobial activity against Candida albicans. The other isolate is the strain Streptomyces griseoaurantiacus AD2, which was isolated from honey, and displays a broad range of antimicrobial activity against different Gram-positive bacteria, including pathogens such as Staphylococcus aureus and Enterococus faecalis. Cultures of S. griseoaurantiacus AD2 have the capacity to produce the antibacterial compounds undecylprodigiosin and manumycin, while those of S. albus AN1 accumulate antifungal compounds such as candicidins and antimycins. Furthermore, genome and dereplication analyses suggest that the number of putative bioactive metabolites produced by AD2 and AN1 is considerably high, including compounds with anti-microbial and anti-cancer properties. Our results postulate that beehives are a promising source for the discovery of novel bioactive compounds that might be of interest to the agri-food sector and healthcare pharmaceuticals.

Toward accurate diagnosis and surveillance of bacterial infections using enhanced strain-level metagenomic next-generation sequencing of infected body fluids

Metagenomic next-generation sequencing (mNGS) enables comprehensive pathogen detection and has become increasingly popular in clinical diagnosis. The distinct pathogenic traits between strains require mNGS to achieve a strain-level resolution, but an equivocal concept of 'strain' as well as the low pathogen loads in most clinical specimens hinders such strain awareness. Here we introduce a metagenomic intra-species typing (MIST) tool (https://github.com/pandafengye/MIST), which hierarchically organizes reference genomes based on average nucleotide identity (ANI) and performs maximum likelihood estimation to infer the strain-level compositional abundance. In silico analysis using synthetic datasets showed that MIST accurately predicted the strain composition at a 99.9% average nucleotide identity (ANI) resolution with a merely 0.001× sequencing depth. When applying MIST on 359 culture-positive and 359 culture-negative real-world specimens of infected body fluids, we found the presence of multiple-strain reached considerable frequencies (30.39%-93.22%), which were otherwise underestimated by current diagnostic techniques due to their limited resolution. Several high-risk clones were identified to be prevalent across samples, including Acinetobacter baumannii sequence type (ST)208/ST195, Staphylococcus aureus ST22/ST398 and Klebsiella pneumoniae ST11/ST15, indicating potential outbreak events occurring in the clinical settings. Interestingly, contaminations caused by the engineered Escherichia coli strain K-12 and BL21 throughout the mNGS datasets were also identified by MIST instead of the statistical decontamination approach. Our study systemically characterized the infected body fluids at the strain level for the first time. Extension of mNGS testing to the strain level can greatly benefit clinical diagnosis of bacterial infections, including the identification of multi-strain infection, decontamination and infection control surveillance.

VirStrain: a strain identification tool for RNA viruses

Viruses change constantly during replication, leading to high intra-species diversity. Although many changes are neutral or deleterious, some can confer on the virus different biological properties such as better adaptability. In addition, viral genotypes often have associated metadata, such as host residence, which can help with inferring viral transmission during pandemics. Thus, subspecies analysis can provide important insights into virus characterization. Here, we present VirStrain, a tool taking short reads as input with viral strain composition as output. We rigorously test VirStrain on multiple simulated and real virus sequencing datasets. VirStrain outperforms the state-of-the-art tools in both sensitivity and accuracy.

Respiratory tract infections and gut microbiome modifications: A systematic review

Respiratory tract infections (RTIs) are extremely common and can cause gastrointestinal tract symptoms and changes to the gut microbiota, yet these effects are poorly understood. We conducted a systematic review to evaluate the reported evidence of gut microbiome alterations in patients with a RTI compared to healthy controls (PROSPERO: CRD42019138853). We systematically searched Medline, Embase, Web of Science, Cochrane and the Clinical Trial Database for studies published between January 2015 and June 2021. Studies were eligible for inclusion if they were human cohorts describing the gut microbiome in patients with an RTI compared to healthy controls and the infection was caused by a viral or bacterial pathogen. Dual data screening and extraction with narrative synthesis was performed. We identified 1,593 articles and assessed 11 full texts for inclusion. Included studies (some nested) reported gut microbiome changes in the context of Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) (n = 5), influenza (H1N1 and H7N9) (n = 2), Tuberculosis (TB) (n = 4), Community-Acquired Pneumonia CAP (n = 2) and recurrent RTIs (rRTI) (n = 1) infections. We found studies of patients with an RTI compared to controls reported a decrease in gut microbiome diversity (Shannon) of 1.45 units (95% CI, 0.15-2.50 [p, <0.0001]) and a lower abundance of taxa (p, 0.0086). Meta-analysis of the Shannon value showed considerable heterogeneity between studies (I2, 94.42). Unbiased analysis displayed as a funnel plot revealed a depletion of Lachnospiraceae, Ruminococcaceae and Ruminococcus and enrichment of Enterococcus. There was an important absence in the lack of cohort studies reporting gut microbiome changes and high heterogeneity between studies may be explained by variations in microbiome methods and confounder effects. Further human cohort studies are needed to understand RTI-induced gut microbiome changes to better understand interplay between microbes and respiratory health.

Identification of isolated or mixed strains from long reads: a challenge met on Streptococcus thermophilus using a MinION sequencer

This study aimed to provide efficient recognition of bacterial strains on personal computers from MinION (Nanopore) long read data. Thanks to the fall in sequencing costs, the identification of bacteria can now proceed by whole genome sequencing. MinION is a fast, but highly error-prone sequencing device and it is a challenge to successfully identify the strain content of unknown simple or complex microbial samples. It is heavily constrained by memory management and fast access to the read and genome fragments. Our strategy involves three steps: indexing of known genomic sequences for a given or several bacterial species; a request process to assign a read to a strain by matching it to the closest reference genomes; and a final step looking for a minimum set of strains that best explains the observed reads. We have applied our method, called ORI, on 77 strains of Streptococcus thermophilus. We worked on several genomic distances and obtained a detailed classification of the strains, together with a criterion that allows merging of what we termed 'sibling' strains, only separated by a few mutations. Overall, isolated strains can be safely recognized from MinION data. For mixtures of several non-sibling strains, results depend on strain abundance.

Large-scale k-mer-based analysis of the informational properties of genomes, comparative genomics and taxonomy

Information theoretic approaches are ubiquitous and effective in a wide variety of bioinformatics applications. In comparative genomics, alignment-free methods, based on short DNA words, or k-mers, are particularly powerful. We evaluated the utility of varying k-mer lengths for genome comparisons by analyzing their sequence space coverage of 5805 genomes in the KEGG GENOME database. In subsequent analyses on four k-mer lengths spanning the relevant range (11, 21, 31, 41), hierarchical clustering of 1634 genus-level representative genomes using pairwise 21- and 31-mer Jaccard similarities best recapitulated a phylogenetic/taxonomic tree of life with clear boundaries for superkingdom domains and high subtree similarity for named taxons at lower levels (family through phylum). By analyzing ~14.2M prokaryotic genome comparisons by their lowest-common-ancestor taxon levels, we detected many potential misclassification errors in a curated database, further demonstrating the need for wide-scale adoption of quantitative taxonomic classifications based on whole-genome similarity.

Genomic Sequence Analysis of Methicillin- and Carbapenem-Resistant Bacteria Isolated from Raw Sewage

Antibiotic resistance is one of the largest threats facing global health. Wastewater treatment plants are well-known hot spots for interaction between diverse bacteria, genetic exchange, and antibiotic resistance. Nonpathogenic bacteria theoretically act as reservoirs of antibiotic resistance subsequently transferring antibiotic resistance genes to pathogens, indicating that evolutionary processes occur outside clinical settings and may drive patterns of drug-resistant infections. We isolated and sequenced 100 bacterial strains from five wastewater treatment plants to analyze regional dynamics of antibiotic resistance in the California Central Valley. The results demonstrate the presence of a wide diversity of pathogenic and nonpathogenic bacteria, with an arithmetic mean of 5.1 resistance genes per isolate. Forty-three percent of resistance genes were located on plasmids, suggesting that large levels of gene transfer between bacteria that otherwise may not co-occur are facilitated by wastewater treatment. One of the strains detected was a Bacillus carrying pX01 and pX02 anthrax-like plasmids and multiple drug resistance genes. A correlation between resistance genes and taxonomy indicates that taxon-specific evolutionary studies may be useful in determining and predicting patterns of antibiotic resistance. Conversely, a lack of geographic correlation may indicate that landscape genetic studies to understand the spread of antibiotic resistance genes should be carried out at broader scales. This large data set provides insights into how pathogenic and nonpathogenic bacteria interact in wastewater environments and the resistance genes which may be horizontally transferred between them. This can help in determining the mechanisms leading to the increasing prevalence of drug-resistant infections observed in clinical settings. IMPORTANCE The reasons for the increasing prevalence of antibiotic-resistant infections are complex and associated with myriad clinical and environmental processes. Wastewater treatment plants operate as nexuses of bacterial interaction and are known hot spots for genetic exchange between bacteria, including antibiotic resistance genes. We isolated and sequenced 100 drug-resistant bacteria from five wastewater treatment plants in California's Central Valley, characterizing widespread gene sharing between pathogens and nonpathogens. We identified a novel, multiresistant Bacillus carrying anthrax-like plasmids. This empirical study supports the likelihood of evolutionary and population processes in the broader environment affecting the prevalence of clinical drug-resistant infections and identifies several taxa that may operate as reservoirs and vectors of antibiotic resistance genes.

Risk Factors for and Mechanisms of COlistin Resistance Among Enterobacterales: Getting at the CORE of the Issue

BACKGROUND

Despite the recent emergence of plasmid-mediated colistin resistance, the epidemiology and mechanisms of colistin-resistant Enterobacterales (CORE) infections remain poorly understood.

METHODS

A case-case-control study was conducted utilizing routine clinical isolates obtained at a single tertiary health system in Ann Arbor, Michigan. Patients with CORE isolates from January 1, 2016, to March 31, 2017, were matched 1:1 with patients with colistin-susceptible Enterobacterales (COSE) and uninfected controls. Multivariable logistic regression was used to compare clinical and microbiologic features of patients with CORE and COSE to controls. A subset of available CORE isolates underwent whole-genome sequencing to identify putative colistin resistance genes.

RESULTS

Of 16 373 tested clinical isolates, 166 (0.99%) were colistin-resistant, representing 103 unique patients. Among 103 CORE isolates, 103 COSE isolates, and 102 uninfected controls, antibiotic exposure in the antecedent 90 days and age >55 years were predictors of both CORE and COSE. Of 33 isolates that underwent whole-genome sequencing, a large variety of mutations associated with colistin resistance were identified, including 4 mcr-1/mcr-1.1 genes and 4 pmrA/B mutations among 9 Escherichia coli isolates and 5 mgrB and 3 PmrA mutations among 8 Klebsiella pneumoniae isolates. Genetic mutations found in Enterobacter species were not associated with known phenotypic colistin resistance.

CONCLUSIONS

Increased age and prior antibiotic receipt were associated with increased risk for patients with CORE and for patients with COSE. Mcr-1, pmrA/B, and mgrB were the predominant colistin resistance-associated mutations identified among E. coli and K. pneumoniae, respectively. Mechanisms of colistin resistance among Enterobacter species could not be determined.

Differences in the genome, methylome, and transcriptome do not differentiate isolates of Streptococcus equi subsp. equi from horses with acute clinical signs from isolates of inapparent carriers

Streptococcus equi subsp. equi (SEE) is a host-restricted bacterium that causes the common infectious upper respiratory disease known as strangles in horses. Perpetuation of SEE infection appears attributable to inapparent carrier horses because it neither persists long-term in the environment nor infects other host mammals or vectors, and infection results in short-lived immunity. Whether pathogen factors enable SEE to remain in horses without causing clinical signs remains poorly understood. Thus, our objective was to use next-generation sequencing technologies to characterize the genome, methylome, and transcriptome of isolates of SEE from horses with acute clinical strangles and inapparent carrier horses—including isolates recovered from individual horses sampled repeatedly—to assess pathogen-associated changes that might reflect specific adaptions of SEE to the host that contribute to inapparent carriage. The accessory genome elements and methylome of SEE isolates from Sweden and Pennsylvania revealed no significant or consistent differences between acute clinical and inapparent carrier isolates of SEE. RNA sequencing of SEE isolates from Pennsylvania demonstrated no genes that were differentially expressed between acute clinical and inapparent carrier isolates of SEE. The absence of specific, consistent changes in the accessory genomes, methylomes, and transcriptomes of acute clinical and inapparent carrier isolates of SEE indicates that adaptations of SEE to the host are unlikely to explain the carrier state of SEE. Efforts to understand the carrier state of SEE should instead focus on host factors.

mixtureS: a novel tool for bacterial strain genome reconstruction from reads

MOTIVATION

It is essential to study bacterial strains in environmental samples. Existing methods and tools often depend on known strains or known variations, cannot work on individual samples, not reliable, or not easy to use, etc. It is thus important to develop more user-friendly tools that can identify bacterial strains more accurately.

RESULTS

We developed a new tool called mixtureS that can de novo identify bacterial strains from shotgun reads of a clonal or metagenomic sample, without prior knowledge about the strains and their variations. Tested on 243 simulated datasets and 195 experimental datasets, mixtureS reliably identified the strains, their numbers and their abundance. Compared with three tools, mixtureS showed better performance in almost all simulated datasets and the vast majority of experimental datasets.

AVAILABILITY AND IMPLEMENTATION

The source code and tool mixtureS is available at http://www.cs.ucf.edu/˜xiaoman/mixtureS/.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Genomics of Staphylococcus aureus ocular isolates

Staphylococcus aureus is a major cause of ocular infections, often resulting in devastating vision loss. Despite the significant morbidity associated with these infections, little is yet known regarding the specific strain types that may have a predilection for ocular tissues nor the set of virulence factors that drive its pathogenicity in this specific biological niche. Whole genome sequencing (WGS) can provide valuable insight in this regard by providing a prospective, comprehensive assessment of the strain types and virulence factors driving disease among specific subsets of clinical isolates. As such, a set of 163-member S. aureus ocular clinical strains were sequenced and assessed for both common strain types (multilocus sequence type (MLST), spa, agr) associated with ocular infections as well as the presence/absence of 235 known virulence factors in a high throughput manner. This ocular strain set was then directly compared to a fully sequenced 116-member non-ocular S. aureus strain set curated from NCBI in order to identify key differences between ocular and non-ocular S. aureus isolates. The most common sequence types found among ocular S. aureus isolates were ST5, ST8 and ST30, generally reflecting circulating non-ocular pathogenic S. aureus strains. However, importantly, ocular isolates were found to be significantly enriched for a set of enterotoxins, suggesting a potential role for this class of virulence factors in promoting ocular disease. Further genomic analysis revealed that these enterotoxins are located on mobile pathogenicity islands, thus horizontal gene transfer may promote the acquisition of enterotoxins, potentially amplifying S. aureus virulence in ocular tissues.

Molecular Characterization of Enterococcus Isolates From Different Sources in Estonia Reveals Potential Transmission of Resistance Genes Among Different Reservoirs

In this study, we aimed to characterize the population structure, drug resistance mechanisms, and virulence genes of Enterococcus isolates in Estonia. Sixty-one Enterococcus faecalis and 34 Enterococcus faecium isolates were collected between 2012 and 2014 across the country from various sites and sources, including farm animals and poultry (n = 53), humans (n = 12), environment (n = 24), and wild birds (n = 44). Clonal relationships of the strains were determined by whole-genome sequencing and analyzed by multi-locus sequence typing. We determined the presence of acquired antimicrobial resistance genes and 23S rRNA mutations, virulence genes, and also the plasmid or chromosomal origin of the genes using dedicated DNA sequence analysis tools available and/or homology search against an ad hoc compiled database of relevant sequences. Two E. faecalis isolates from human with vanB genes were highly resistant to vancomycin. Closely related E. faecalis strains were isolated from different host species. This indicates interspecies spread of strains and potential transfer of antibiotic resistance. Genomic context analysis of the resistance genes indicated frequent association with plasmids and mobile genetic elements. Resistance genes are often present in the identical genetic context in strains with diverse origins, suggesting the occurrence of transfer events.

Computational Methods for Strain-Level Microbial Detection in Colony and Metagenome Sequencing Data

Metagenomic sequencing is a powerful tool for examining the diversity and complexity of microbial communities. Most widely used tools for taxonomic profiling of metagenomic sequence data allow for a species-level overview of the composition. However, individual strains within a species can differ greatly in key genotypic and phenotypic characteristics, such as drug resistance, virulence and growth rate. Therefore, the ability to resolve microbial communities down to the level of individual strains within a species is critical to interpreting metagenomic data for clinical and environmental applications, where identifying a particular strain, or tracking a particular strain across a set of samples, can help aid in clinical diagnosis and treatment, or in characterizing yet unstudied strains across novel environmental locations. Recently published approaches have begun to tackle the problem of resolving strains within a particular species in metagenomic samples. In this review, we present an overview of these new algorithms and their uses, including methods based on assembly reconstruction and methods operating with or without a reference database. While existing metagenomic analysis methods show reasonable performance at the species and higher taxonomic levels, identifying closely related strains within a species presents a bigger challenge, due to the diversity of databases, genetic relatedness, and goals when conducting these analyses. Selection of which metagenomic tool to employ for a specific application should be performed on a case-by case basis as these tools have strengths and weaknesses that affect their performance on specific tasks. A comprehensive benchmark across different use case scenarios is vital to validate performance of these tools on microbial samples. Because strain-level metagenomic analysis is still in its infancy, development of more fine-grained, high-resolution algorithms will continue to be in demand for the future.

Method for the Identification of Plant DNA in Food Using Alignment-Free Analysis of Sequencing Reads: A Case Study on Lupin

Fast and reliable analytical methods for the identification of plants from metagenomic samples play an important role in identifying the components of complex mixtures of processed biological materials, including food, herbal products, gut contents or environmental samples. Different PCR-based methods that are commonly used for plant identification from metagenomic samples are often inapplicable due to DNA degradation, a low level of successful amplification or a lack of detection power. We introduce a method that combines metagenomic sequencing and an alignment-free k-mer based approach for the identification of plant DNA in processed metagenomic samples. Our method identifies plant DNA directly from metagenomic sequencing reads and does not require mapping or assembly of the reads. We identified more than 31,000 Lupinus-specific 32-mers from assembled chloroplast genome sequences. We demonstrate that lupin DNA can be detected from controlled mixtures of sequences from target species (different Lupinus species) and closely related non-target species (Arachis hypogaea, Glycine max, Pisum sativum, Vicia faba, Phaseolus vulgaris, Lens culinaris, and Cicer arietinum). Moreover, these 32-mers are detectable in the following processed samples: lupin flour, conserved seeds and baked cookies containing different amounts of lupin flour. Under controlled conditions, lupin-specific components are detectable in baked cookies containing a minimum of 0.05% of lupin flour in wheat flour.

QuantTB - a method to classify mixed Mycobacterium tuberculosis infections within whole genome sequencing data

BACKGROUND

Mixed infections of Mycobacterium tuberculosis and antibiotic heteroresistance continue to complicate tuberculosis (TB) diagnosis and treatment. Detection of mixed infections has been limited to molecular genotyping techniques, which lack the sensitivity and resolution to accurately estimate the multiplicity of TB infections. In contrast, whole genome sequencing offers sensitive views of the genetic differences between strains of M. tuberculosis within a sample. Although metagenomic tools exist to classify strains in a metagenomic sample, most tools have been developed for more divergent species, and therefore cannot provide the sensitivity required to disentangle strains within closely related bacterial species such as M. tuberculosis. Here we present QuantTB, a method to identify and quantify individual M. tuberculosis strains in whole genome sequencing data. QuantTB uses SNP markers to determine the combination of strains that best explain the allelic variation observed in a sample. QuantTB outputs a list of identified strains, their corresponding relative abundances, and a list of drugs for which resistance-conferring mutations (or heteroresistance) have been predicted within the sample.

RESULTS

We show that QuantTB has a high degree of resolution and is capable of differentiating communities differing by less than 25 SNPs and identifying strains down to 1× coverage. Using simulated data, we found QuantTB outperformed other metagenomic strain identification tools at detecting strains and quantifying strain multiplicity. In a real-world scenario, using a dataset of 50 paired clinical isolates from a study of patients with either reinfections or relapses, we found that QuantTB could detect mixed infections and reinfections at rates concordant with a manually curated approach.

CONCLUSION

QuantTB can determine infection multiplicity, identify hetero-resistance patterns, enable differentiation between relapse and re-infection, and clarify transmission events across seemingly unrelated patients - even in low-coverage (1×) samples. QuantTB outperforms existing tools and promises to serve as a valuable resource for both clinicians and researchers working with clinical TB samples.

Kmer-db: instant evolutionary distance estimation

Summary

Kmer-db is a new tool for estimating evolutionary relationship on the basis of k-mers extracted from genomes or sequencing reads. Thanks to an efficient data structure and parallel implementation, our software estimates distances between 40 715 pathogens in <7 min (on a modern workstation), 26 times faster than Mash, its main competitor.

Availability and implementation

https://github.com/refresh-bio/kmer-db and http://sun.aei.polsl.pl/REFRESH/kmer-db.

Supplementary information

Supplementary data are available at Bioinformatics online.

SEPATH: benchmarking the search for pathogens in human tissue whole genome sequence data leads to template pipelines

BACKGROUND

Human tissue is increasingly being whole genome sequenced as we transition into an era of genomic medicine. With this arises the potential to detect sequences originating from microorganisms, including pathogens amid the plethora of human sequencing reads. In cancer research, the tumorigenic ability of pathogens is being recognized, for example, Helicobacter pylori and human papillomavirus in the cases of gastric non-cardia and cervical carcinomas, respectively. As of yet, no benchmark has been carried out on the performance of computational approaches for bacterial and viral detection within host-dominated sequence data.

RESULTS

We present the results of benchmarking over 70 distinct combinations of tools and parameters on 100 simulated cancer datasets spiked with realistic proportions of bacteria. mOTUs2 and Kraken are the highest performing individual tools achieving median genus-level F1 scores of 0.90 and 0.91, respectively. mOTUs2 demonstrates a high performance in estimating bacterial proportions. Employing Kraken on unassembled sequencing reads produces a good but variable performance depending on post-classification filtering parameters. These approaches are investigated on a selection of cervical and gastric cancer whole genome sequences where Alphapapillomavirus and Helicobacter are detected in addition to a variety of other interesting genera.

CONCLUSIONS

We provide the top-performing pipelines from this benchmark in a unifying tool called SEPATH, which is amenable to high throughput sequencing studies across a range of high-performance computing clusters. SEPATH provides a benchmarked and convenient approach to detect pathogens in tissue sequence data helping to determine the relationship between metagenomics and disease.

Machine Learning Approaches for Epidemiological Investigations of Food-Borne Disease Outbreaks

Foodborne diseases (FBDs) are infections of the gastrointestinal tract caused by foodborne pathogens (FBPs) such as bacteria [Salmonella, Listeria monocytogenes and Shiga toxin-producing E. coli (STEC)] and several viruses, but also parasites and some fungi. Artificial intelligence (AI) and its sub-discipline machine learning (ML) are re-emerging and gaining an ever increasing popularity in the scientific community and industry, and could lead to actionable knowledge in diverse ranges of sectors including epidemiological investigations of FBD outbreaks and antimicrobial resistance (AMR). As genotyping using whole-genome sequencing (WGS) is becoming more accessible and affordable, it is increasingly used as a routine tool for the detection of pathogens, and has the potential to differentiate between outbreak strains that are closely related, identify virulence/resistance genes and provide improved understanding of transmission events within hours to days. In most cases, the computational pipeline of WGS data analysis can be divided into four (though, not necessarily consecutive) major steps: de novo genome assembly, genome characterization, comparative genomics, and inference of phylogeny or phylogenomics. In each step, ML could be used to increase the speed and potentially the accuracy (provided increasing amounts of high-quality input data) of identification of the source of ongoing outbreaks, leading to more efficient treatment and prevention of additional cases. In this review, we explore whether ML or any other form of AI algorithms have already been proposed for the respective tasks and compare those with mechanistic model-based approaches.

Lactobacilli Isolated From Wild Boar (Sus scrofa) Antagonize Mycobacterium bovis Bacille Calmette-Guerin (BCG) in a Species-Dependent Manner

Background: Wildlife poses a significant burden for the complete eradication of bovine tuberculosis (bTB). In particular, wild boar (Sus scrofa) is one of the most important reservoirs of Mycobacterium bovis, the causal agent of bTB. Wild boar can display from mild TB lesions, usually found in head lymph nodes, to generalized TB lesions distributed in different anatomical regions; but rarely clinical signs, which complicates the diagnosis of Mycobacterium bovis infection and bTB control. Among the possibilities for this variability in lesion distribution is the influence of the host-beneficial commensal-primed immune barrier. In this respect, beneficial microbes may delay bTB dissemination as a consequence of an antagonistic competition for nutrients and phagocytes. In order to explore this possibility, we have tested whether typical commensals such as lactobacilli have the capacity to reduce the survival rate of the surrogate M. bovis strain Bacillus Calmette-Guerin (BCG); and to modulate its phagocyte intake.

Results: Three Lactobacillus species, L. casei, L. plantarum, and L. salivarius, isolated from wild boar feces displayed a pH-dependent inhibitory activity against BCG and influenced its intake by porcine blood phagocytes in a species-dependent manner. All lactobacilli showed a very significant bactericidal effect against BCG at low pH, but only isolates of L. plantarum and L. casei displayed such antimycobacterial activity at neutral pH. The genomes of these isolates revealed the presence of two-peptide bacteriocins whose precursor genes up-regulate in the presence of BCG cells. Furthermore, L. plantarum reduced significantly the BCG phagocytic intake, whereas L. casei had the opposite effect. L. salivarius had no significant influence on the phagocytic response to BCG.

Conclusions: Our in vitro results show that lactobacilli isolated from wild boar antagonize BCG as a consequence of their antimycobacterial activity and a competitive phagocytic response. These findings suggest that commensal bacteria could play a beneficial role in influencing the outcome of bTB dissemination. Further work with lactobacilli as a potential competitive pressure to control bTB will need to take into account the complex nature of the commensal microbiome, the specific immunity of the wild boar and the in vivo infection context with pathogenic strains of M. bovis.

UltraStrain: An NGS-Based Ultra Sensitive Strain Typing Method for Salmonella enterica

In the last few years, advances in next-generation sequencing (NGS) technology for whole genome sequencing (WGS) of foodborne pathogens have provided drastic improvements in food pathogen outbreak surveillance. WGS of foodborne pathogen enables identification of pathogens from food or environmental samples, including difficult-to-detect pathogens in culture-negative infections. Compared to traditional low-resolution methods such as the pulsed-field gel electrophoresis (PFGE), WGS provides advantages to differentiate even closely related strains of the same species, thus enables rapid identification of food-source associated with pathogen outbreak events for a fast mitigation plan. In this paper, we present UltraStrain, which is a fast and ultra sensitive pathogen detection and strain typing method for Salmonella enterica (S. enterica) based on WGS data analysis. In the proposed method, a noise filtering step is first performed where the raw sequencing data are mapped to a synthetic species-specific reference genome generated from S. enterica specific marker sequences to avoid potential interference from closely related species for low spike samples. After that, a statistical learning based method is used to identify candidate strains, from a database of known S. enterica strains, that best explain the retained S. enterica specific reads.Finally, a refinement step is further performed by mapping all the reads before filtering onto the identified top candidate strains, and recalculating the probability of presence for each candidate strain. Experiment results using both synthetic and real sequencing data show that the proposed method is able to identify the correct S. enterica strains from low-spike samples, and outperforms several existing strain-typing methods in terms of sensitivity and accuracy.

Exploring Foodborne Pathogen Ecology and Antimicrobial Resistance in the Light of Shotgun Metagenomics

In this chapter, applications of shotgun metagenomics for taxonomic profiling and functional investigation of food microbial communities with a focus on antimicrobial resistance (AMR) were overviewed in the light of last data in the field. Potentialities of metagenomic approach, along with the challenges encountered for a wider and routinely use in food safety was discussed.

PlasmidSeeker: identification of known plasmids from bacterial whole genome sequencing reads

Background

Plasmids play an important role in the dissemination of antibiotic resistance, making their detection an important task. Using whole genome sequencing (WGS), it is possible to capture both bacterial and plasmid sequence data, but short read lengths make plasmid detection a complex problem.

Results

We developed a tool named PlasmidSeeker that enables the detection of plasmids from bacterial WGS data without read assembly. The PlasmidSeeker algorithm is based on k-mers and uses k-mer abundance to distinguish between plasmid and bacterial sequences. We tested the performance of PlasmidSeeker on a set of simulated and real bacterial WGS samples, resulting in 100% sensitivity and 99.98% specificity.

Conclusion

PlasmidSeeker enables quick detection of known plasmids and complements existing tools that assemble plasmids de novo. The PlasmidSeeker source code is stored on GitHub: https://github.com/bioinfo-ut/PlasmidSeeker.

Method for the Identification of Taxon-Specific k-mers from Chloroplast Genome: A Case Study on Tomato Plant (Solanum lycopersicum)

Polymerase chain reaction and different barcoding methods commonly used for plant identification from metagenomics samples are based on the amplification of a limited number of pre-selected barcoding regions. These methods are often inapplicable due to DNA degradation, low amplification success or low species discriminative power of selected genomic regions. Here we introduce a method for the rapid identification of plant taxon-specific k-mers, that is applicable for the fast detection of plant taxa directly from raw sequencing reads without aligning, mapping or assembling the reads. We identified more than 800 Solanum lycopersicum specific k-mers (32 nucleotides in length) from 42 different chloroplast genome regions using the developed method. We demonstrated that identified k-mers are also detectable in whole genome sequencing raw reads from S. lycopersicum. Also, we demonstrated the usability of taxon-specific k-mers in artificial mixtures of sequences from closely related species. Developed method offers a novel strategy for fast identification of taxon-specific genome regions and offers new perspectives for detection of plant taxa directly from sequencing raw reads.

MentaLiST - A fast MLST caller for large MLST schemes

MLST (multi-locus sequence typing) is a classic technique for genotyping bacteria, widely applied for pathogen outbreak surveillance. Traditionally, MLST is based on identifying sequence types from a small number of housekeeping genes. With the increasing availability of whole-genome sequencing data, MLST methods have evolved towards larger typing schemes, based on a few hundred genes [core genome MLST (cgMLST)] to a few thousand genes [whole genome MLST (wgMLST)]. Such large-scale MLST schemes have been shown to provide a finer resolution and are increasingly used in various contexts such as hospital outbreaks or foodborne pathogen outbreaks. This methodological shift raises new computational challenges, especially given the large size of the schemes involved. Very few available MLST callers are currently capable of dealing with large MLST schemes. We introduce MentaLiST, a new MLST caller, based on a k-mer voting algorithm and written in the Julia language, specifically designed and implemented to handle large typing schemes. We test it on real and simulated data to show that MentaLiST is faster than any other available MLST caller while providing the same or better accuracy, and is capable of dealing with MLST schemes with up to thousands of genes while requiring limited computational resources. MentaLiST source code and easy installation instructions using a Conda package are available at https://github.com/WGS-TB/MentaLiST.

Alignment-free sequence comparison: benefits, applications, and tools

Alignment-free sequence analyses have been applied to problems ranging from whole-genome phylogeny to the classification of protein families, identification of horizontally transferred genes, and detection of recombined sequences. The strength of these methods makes them particularly useful for next-generation sequencing data processing and analysis. However, many researchers are unclear about how these methods work, how they compare to alignment-based methods, and what their potential is for use for their research. We address these questions and provide a guide to the currently available alignment-free sequence analysis tools.

FastGT: an alignment-free method for calling common SNVs directly from raw sequencing reads

We have developed a computational method that counts the frequencies of unique k-mers in FASTQ-formatted genome data and uses this information to infer the genotypes of known variants. FastGT can detect the variants in a 30x genome in less than 1 hour using ordinary low-cost server hardware. The overall concordance with the genotypes of two Illumina “Platinum” genomes is 99.96%, and the concordance with the genotypes of the Illumina HumanOmniExpress is 99.82%. Our method provides k-mer database that can be used for the simultaneous genotyping of approximately 30 million single nucleotide variants (SNVs), including >23,000 SNVs from Y chromosome. The source code of FastGT software is available at GitHub (https://github.com/bioinfo-ut/GenomeTester4/).