Noelle R. Noyes

MEGARes: an antimicrobial resistance database for high throughput sequencing

Antimicrobial resistance has become an imminent concern for public health. As methods for detection and characterization of antimicrobial resistance move from targeted culture and polymerase chain reaction to high throughput metagenomics, appropriate resources for the analysis of large-scale data are required. Currently, antimicrobial resistance databases are tailored to smaller-scale, functional profiling of genes using highly descriptive annotations. Such characteristics do not facilitate the analysis of large-scale, ecological sequence datasets such as those produced with the use of metagenomics for surveillance. In order to overcome these limitations, we present MEGARes (https://megares.meglab.org), a hand-curated antimicrobial resistance database and annotation structure that provides a foundation for the development of high throughput acyclical classifiers and hierarchical statistical analysis of big data. MEGARes can be browsed as a stand-alone resource through the website or can be easily integrated into sequence analysis pipelines through download. Also via the website, we provide documentation for AmrPlusPlus, a user-friendly Galaxy pipeline for the analysis of high throughput sequencing data that is pre-packaged for use with the MEGARes database.

Use of Metagenomic Shotgun Sequencing Technology To Detect Foodborne Pathogens within the Microbiome of the Beef Production Chain

ABSTRACT Foodborne illnesses associated with pathogenic bacteria are a global public health and economic challenge. The diversity of microorganisms (pathogenic and nonpathogenic) that exists within the food and meat industries complicates efforts to understand pathogen ecology. Further, little is known about the interaction of pathogens within the microbiome throughout the meat production chain. Here, a metagenomic approach and shotgun sequencing technology were used as tools to detect pathogenic bacteria in environmental samples collected from the same groups of cattle at different longitudinal processing steps of the beef production chain: cattle entry to feedlot, exit from feedlot, cattle transport trucks, abattoir holding pens, and the end of the fabrication system. The log read counts classified as pathogens per million reads for Salmonella enterica, Listeria monocytogenes, Escherichia coli, Staphylococcus aureus, Clostridium spp. (C. botulinum and C. perfringens), and Campylobacter spp. (C. jejuni, C. coli, and C. fetus) decreased over subsequential processing steps. Furthermore, the normalized read counts for S. enterica, E. coli, and C. botulinum were greater in the final product than at the feedlots, indicating that the proportion of these bacteria increased (the effect on absolute numbers was unknown) within the remaining microbiome. From an ecological perspective, data indicated that shotgun metagenomics can be used to evaluate not only the microbiome but also shifts in pathogen populations during beef production. Nonetheless, there were several challenges in this analysis approach, one of the main ones being the identification of the specific pathogen from which the sequence reads originated, which makes this approach impractical for use in pathogen identification for regulatory and confirmation purposes.

Resistome diversity in cattle and the environment decreases during beef production

Antimicrobial resistant determinants (ARDs) can be transmitted from livestock systems through meat products or environmental effluents. The public health risk posed by these two routes is not well understood, particularly in non-pathogenic bacteria. We collected pooled samples from 8 groups of 1741 commercial cattle as they moved through the process of beef production from feedlot entry through slaughter. We recorded antimicrobial drug exposures and interrogated the resistome at points in production when management procedures could potentially influence ARD abundance and/or transmission. Over 300 unique ARDs were identified. Resistome diversity decreased while cattle were in the feedlot, indicating selective pressure. ARDs were not identified in beef products, suggesting that slaughter interventions may reduce the risk of transmission of ARDs to beef consumers. This report highlights the utility and limitations of metagenomics for assessing public health risks regarding antimicrobial resistance, and demonstrates that environmental pathways may represent a greater risk than the food supply. DOI: http://dx.doi.org/10.7554/eLife.13195.001

Characterization of the resistome in manure, soil and wastewater from dairy and beef production systems.

It has been proposed that livestock production effluents such as wastewater, airborne dust and manure increase the density of antimicrobial resistant bacteria and genes in the environment. The public health risk posed by this proposed outcome has been difficult to quantify using traditional microbiological approaches. We utilized shotgun metagenomics to provide a first description of the resistome of North American dairy and beef production effluents, and identify factors that significantly impact this resistome. We identified 34 mechanisms of antimicrobial drug resistance within 34 soil, manure and wastewater samples from feedlot, ranch and dairy operations. The majority of resistance-associated sequences found in all samples belonged to tetracycline resistance mechanisms. We found that the ranch samples contained significantly fewer resistance mechanisms than dairy and feedlot samples, and that the resistome of dairy operations differed significantly from that of feedlots. The resistome in soil, manure and wastewater differed, suggesting that management of these effluents should be tailored appropriately. By providing a baseline of the cattle production waste resistome, this study represents a solid foundation for future efforts to characterize and quantify the public health risk posed by livestock effluents.

Mannheimia haemolytica in Feedlot Cattle: Prevalence of Recovery and Associations with Antimicrobial Use, Resistance, and Health Outcomes

Background Mannheimia haemolytica is an important etiological agent in bovine respiratory disease. Objectives Explore risk factors for recovery of susceptible and resistant M. haemolytica in feedlot cattle and explore associations with health outcomes. Animals Cattle (n = 5,498) from 4 feedlots sampled at arrival and later in feeding period. Methods Susceptibility of M. haemolytica isolates tested for 21 antimicrobials. Records of antimicrobial use and health events analyzed using multivariable regression. Results M. haemolytica recovered from 29% of cattle (1,596/5,498), 13.1% at arrival (95% CI, 12.3–14.1%), and 19.8% at second sampling (95% CI, 18.7–20.9%). Nearly half of study cattle received antimicrobial drugs (AMDs) parenterally, mostly as metaphylactic treatment at arrival. Individual parenteral AMD exposures were associated with decreased recovery of M. haemolytica (OR, 0.2; 95% CI, 0.02–1.2), whereas exposure in penmates was associated with increased recovery (OR, 1.5; 95% CI, 1.05–2.2). Most isolates were pan‐susceptible (87.8%; 95% CI, 87.0–89.4%). AMD exposures were not associated with resistance to any single drug. Multiply‐resistant isolates were rare (5.9%; 95% CI, 5.1–6.9%), but AMD exposures in pen mates were associated with increased odds of recovering multiply‐resistant M. haemolytica (OR, 23.9; 95% CI, 8.4–68.3). Cattle positive for M. haemolytica on arrival were more likely to become ill within 10 days (OR, 1.7; 95% CI, 1.1–2.4). Conclusions and Clinical Importance Resistance generally was rare in M. haemolytica. Antimicrobial drug exposures in penmates increased the risk of isolating susceptible and multiply‐resistant M. haemolytica, a finding that could be explained by contagious spread.

Succinct colored de Bruijn graphs

Motivation In 2012, Iqbal et al. introduced the colored de Bruijn graph, a variant of the classic de Bruijn graph, which is aimed at ‘detecting and genotyping simple and complex genetic variants in an individual or population’. Because they are intended to be applied to massive population level data, it is essential that the graphs be represented efficiently. Unfortunately, current succinct de Bruijn graph representations are not directly applicable to the colored de Bruijn graph, which requires additional information to be succinctly encoded as well as support for non‐standard traversal operations. Results Our data structure dramatically reduces the amount of memory required to store and use the colored de Bruijn graph, with some penalty to runtime, allowing it to be applied in much larger and more ambitious sequence projects than was previously possible. Availability and Implementation https://github.com/cosmo‐team/cosmo/tree/VARI Contact martin.muggli@colostate.edu Supplementary information Supplementary data are available at Bioinformatics online.

Antimicrobial Resistance in Escherichia coli Recovered from Feedlot Cattle and Associations with Antimicrobial Use

The objectives of this study were to estimate the prevalence of antimicrobial resistance (AMR) and to investigate the associations between exposures to antimicrobial drugs (AMDs) and AMR in fecal non-type specific Escherichia coli (NTSEC) recovered from a large population of feedlot cattle. Two-stage random sampling was used to select individually identified cattle for enrollment, which were sampled at arrival and then a second time later in the feeding period. Advanced regression techniques were used to estimate resistance prevalences, and to investigate associations between AMD exposures in enrolled cattle and penmates and AMR identified in NTSEC recovered from the second sample set. Resistance was most commonly detected to tetracycline, streptomycin, and sulfisoxazole, and was rarely identified for critically important AMDs. All cattle were exposed to AMDs in feed, and 45% were treated parenterally. While resistance prevalence generally increased during the feeding period, most AMD exposures were not significantly associated with AMR outcomes. Exposures of enrolled cattle to tetracycline were associated with increased resistance to tetracycline and trimethoprim sulfa, while beta-lactam exposures were associated with decreased likelihood of detecting streptomycin resistance. Pen-level AMD exposure measures were not associated with resistance outcomes. These findings suggest that tetracycline treatment of feedlot cattle can be associated with modest increases in risk for recovery of resistant NTSEC, but the numerous treatments with an advanced macrolide (tulathromycin) were not associated with detectable increases in resistance in NTSEC. All cattle were exposed to in-feed treatments of tetracycline and this could limit the ability to identify the full impact of these exposures, but these exposures varied for enrolled cattle varied, providing an opportunity to evaluate a dose response. While AMD exposures were not associated with detectably increased risks for resistance to critically important AMDs, rare resistance outcomes and infrequent exposure to other important AMDs (e.g., cephalosporins) limited our ability to rigorously investigate questions regarding factors that can influence resistance to these important AMDs.

Associations between sexual habits, menstrual hygiene practices, demographics and the vaginal microbiome as revealed by Bayesian network analysis

The vaginal microbiome plays an influential role in several disease states in reproductive age women, including bacterial vaginosis (BV). While demographic characteristics are associated with differences in vaginal microbiome community structure, little is known about the influence of sexual and hygiene habits. Furthermore, associations between the vaginal microbiome and risk symptoms of bacterial vaginosis have not been fully elucidated. Using Bayesian network (BN) analysis of 16S rRNA gene sequence results, demographic and extensive questionnaire data, we describe both novel and previously documented associations between habits of women and their vaginal microbiome. The BN analysis approach shows promise in uncovering complex associations between disparate data types. Our findings based on this approach support published associations between specific microbiome members (e.g., Eggerthella, Gardnerella, Dialister, Sneathia and Ruminococcaceae), the Nugent score (a BV diagnostic) and vaginal pH (a risk symptom of BV). Additionally, we found that several microbiome members were directly connected to other risk symptoms of BV (such as vaginal discharge, odor, itch, irritation, and yeast infection) including L. jensenii, Corynebacteria, and Proteobacteria. No direct connections were found between the Nugent Score and risk symptoms of BV other than pH, indicating that the Nugent Score may not be the most useful criteria for assessment of clinical BV. We also found that demographics (i.e., age, ethnicity, previous pregnancy) were associated with the presence/absence of specific vaginal microbes. The resulting BN revealed several as-yet undocumented associations between birth control usage, menstrual hygiene practices and specific microbiome members. Many of these complex relationships were not identified using common analytical methods, i.e., ordination and PERMANOVA. While these associations require confirmatory follow-up study, our findings strongly suggest that future studies of the vaginal microbiome and vaginal pathologies should include detailed surveys of participants’ sanitary, sexual and birth control habits, as these can act as confounders in the relationship between the microbiome and disease. Although the BN approach is powerful in revealing complex associations within multidimensional datasets, the need in some cases to discretize the data for use in BN analysis can result in loss of information. Future research is required to alleviate such limitations in constructing BN networks. Large sample sizes are also required in order to allow for the incorporation of a large number of variables (nodes) into the BN, particularly when studying associations between metadata and the microbiome. We believe that this approach is of great value, complementing other methods, to further our understanding of complex associations characteristic of microbiome research.

Modelling considerations in the analysis of associations between antimicrobial use and resistance in beef feedlot cattle

A number of sophisticated modelling approaches are available to investigate potential associations between antimicrobial use (AMU) and resistance (AMR) in animal health settings. All have their advantages and disadvantages, making it unclear as to which model is most appropriate. We used advanced regression modelling to investigate AMU-AMR associations in faecal non-type-specific Escherichia coli (NTSEC) isolates recovered from 275 pens of feedlot cattle. Ten modelling strategies were employed to investigate AMU associations with resistance to chloramphenicol, ampicillin, sulfisoxazole, tetracycline and streptomycin. Goodness-of-fit statistics did not show a consistent advantage for any one model type. Three AMU-AMR associations were significant in all models. Recent parenteral tetracycline use increased the odds of finding tetracycline-resistant NTSEC [odds ratios (OR) 1·1-3·2]; recent parenteral sulfonamide use increased the odds of finding sulfisoxazole-resistant NTSEC (OR 1·4-2·5); and recent parenteral macrolide use decreased the odds of recovering ampicillin-resistant NTSEC (OR 0·03-0·2). Other results varied markedly depending on the modelling approach, emphasizing the importance of exploring and reporting multiple modelling methods based on a balanced consideration of important factors such as study design, mathematical appropriateness, research question and target audience.

Enrichment allows identification of diverse, rare elements in metagenomic resistome-virulome sequencing

BackgroundShotgun metagenomic sequencing is increasingly utilized as a tool to evaluate ecological-level dynamics of antimicrobial resistance and virulence, in conjunction with microbiome analysis. Interest in use of this method for environmental surveillance of antimicrobial resistance and pathogenic microorganisms is also increasing. In published metagenomic datasets, the total of all resistance- and virulence-related sequences accounts for < 1% of all sequenced DNA, leading to limitations in detection of low-abundance resistome-virulome elements. This study describes the extent and composition of the low-abundance portion of the resistome-virulome, using a bait-capture and enrichment system that incorporates unique molecular indices to count DNA molecules and correct for enrichment bias.ResultsThe use of the bait-capture and enrichment system significantly increased on-target sequencing of the resistome-virulome, enabling detection of an additional 1441 gene accessions and revealing a low-abundance portion of the resistome-virulome that was more diverse and compositionally different than that detected by more traditional metagenomic assays. The low-abundance portion of the resistome-virulome also contained resistance genes with public health importance, such as extended-spectrum betalactamases, that were not detected using traditional shotgun metagenomic sequencing. In addition, the use of the bait-capture and enrichment system enabled identification of rare resistance gene haplotypes that were used to discriminate between sample origins.ConclusionsThese results demonstrate that the rare resistome-virulome contains valuable and unique information that can be utilized for both surveillance and population genetic investigations of resistance. Access to the rare resistome-virulome using the bait-capture and enrichment system validated in this study can greatly advance our understanding of microbiome-resistome dynamics.