Zoonotic Source Attribution of Salmonella enterica Serotype Typhimurium Using Genomic Surveillance Data, United States

Bayesian evaluation of meat juice ELISA for detecting Salmonella in slaughtered pigs without specifying a cut-off

BACKGROUND

Consumption of pork and pork products is a major source of human infection with Salmonella. Salmonella is typically subclinical in pigs, making it difficult to identify infected pigs. Therefore, effective surveillance of Salmonella in pigs critically relies on good knowledge on how well the diagnostic tests used perform. A test that has been used in several countries for Salmonella monitoring is serological testing of meat juice using an ELISA (MJ ELISA) to detect antibodies against Salmonella. This MJ ELISA data could be used to estimate infection prevalence and trends. However, as the MJ ELISA output is a sample-to-positive (S/P) ratio, which is a continuous outcome rather than a binary (positive/negative) result, the interpretation of this data depends upon a chosen cut-off.

AIM

To apply Bayesian latent class models (BLCMs) to estimate diagnostic accuracy of the MJ ELISA test values in the absence of a gold standard without needing to apply a cut-off.

METHODS AND RESULTS

BLCMs were fitted to data from a UK abattoir survey carried out in 2006 in order to estimate the diagnostic accuracy of MJ ELISA with respect to the prevalence of active Salmonella infection. This survey consisted of a MJ ELISA applied in parallel with the bacteriological testing of caecal contents, carcass swabs and lymph nodes (n = 625). A BLCM was also fitted to the same data but with dichotomisation of the MJ ELISA results, in order to compare with the model using continuous outcomes. Estimates were obtained for sensitivity and specificity of the ELISA over a range of S/P values and for the bacteriological tests and were found to be similar between the models using continuous and dichotomous ELISA outcomes.

CONCLUSION

The Bayesian method without specifying a cut-off does allow prevalence to be inferred without specifying a cut-off for the ELISA. The study results will be useful for estimating infection prevalence from serological surveillance data.

Development and validation of a random forest algorithm for source attribution of animal and human Salmonella Typhimurium and monophasic variants of S. Typhimurium isolates in England and Wales utilising whole genome sequencing data

Source attribution has traditionally involved combining epidemiological data with different pathogen characterisation methods, including 7-gene multi locus sequence typing (MLST) or serotyping, however, these approaches have limited resolution. In contrast, whole genome sequencing data provide an overview of the whole genome that can be used by attribution algorithms. Here, we applied a random forest (RF) algorithm to predict the primary sources of human clinical Salmonella Typhimurium (S. Typhimurium) and monophasic variants (monophasic S. Typhimurium) isolates. To this end, we utilised single nucleotide polymorphism diversity in the core genome MLST alleles obtained from 1,061 laboratory-confirmed human and animal S. Typhimurium and monophasic S. Typhimurium isolates as inputs into a RF model. The algorithm was used for supervised learning to classify 399 animal S. Typhimurium and monophasic S. Typhimurium isolates into one of eight distinct primary source classes comprising common livestock and pet animal species: cattle, pigs, sheep, other mammals (pets: mostly dogs and horses), broilers, layers, turkeys, and game birds (pheasants, quail, and pigeons). When applied to the training set animal isolates, model accuracy was 0.929 and kappa 0.905, whereas for the test set animal isolates, for which the primary source class information was withheld from the model, the accuracy was 0.779 and kappa 0.700. Subsequently, the model was applied to assign 662 human clinical cases to the eight primary source classes. In the dataset, 60/399 (15.0%) of the animal and 141/662 (21.3%) of the human isolates were associated with a known outbreak of S. Typhimurium definitive type (DT) 104. All but two of the 141 DT104 outbreak linked human isolates were correctly attributed by the model to the primary source classes identified as the origin of the DT104 outbreak. A model that was run without the clonal DT104 animal isolates produced largely congruent outputs (training set accuracy 0.989 and kappa 0.985; test set accuracy 0.781 and kappa 0.663). Overall, our results show that RF offers considerable promise as a suitable methodology for epidemiological tracking and source attribution for foodborne pathogens.

A Review on Microbiological Source Attribution Methods of Human Salmonellosis: From Subtyping to Whole-Genome Sequencing

Salmonella is one of the main causes of human foodborne illness. It is endemic worldwide, with different animals and animal-based food products as reservoirs and vehicles of infection. Identifying animal reservoirs and potential transmission pathways of Salmonella is essential for prevention and control. There are many approaches for source attribution, each using different statistical models and data streams. Some aim to identify the animal reservoir, while others aim to determine the point at which exposure occurred. With the advance of whole-genome sequencing (WGS) technologies, new source attribution models will greatly benefit from the discriminating power gained with WGS. This review discusses some key source attribution methods and their mathematical and statistical tools. We also highlight recent studies utilizing WGS for source attribution and discuss open questions and challenges in developing new WGS methods. We aim to provide a better understanding of the current state of these methodologies with application to Salmonella and other foodborne pathogens that are common sources of illness in the poultry and human sectors.

The advantage of intergenic regions as genomic features for machine-learning-based host attribution of Salmonella Typhimurium from the USA

Salmonella enterica is a taxonomically diverse pathogen with over 2600 serovars associated with a wide variety of animal hosts including humans, other mammals, birds and reptiles. Some serovars are host-specific or host-restricted and cause disease in distinct host species, while others, such as serovar S. Typhimurium (STm), are generalists and have the potential to colonize a wide variety of species. However, even within generalist serovars such as STm it is becoming clear that pathovariants exist that differ in tropism and virulence. Identifying the genetic factors underlying host specificity is complex, but the availability of thousands of genome sequences and advances in machine learning have made it possible to build specific host prediction models to aid outbreak control and predict the human pathogenic potential of isolates from animals and other reservoirs. We have advanced this area by building host-association prediction models trained on a wide range of genomic features and compared them with predictions based on nearest-neighbour phylogeny. SNPs, protein variants (PVs), antimicrobial resistance (AMR) profiles and intergenic regions (IGRs) were extracted from 3883 high-quality STm assemblies collected from humans, swine, bovine and poultry in the USA, and used to construct Random Forest (RF) machine learning models. An additional 244 recent STm assemblies from farm animals were used as a test set for further validation. The models based on PVs and IGRs had the best performance in terms of predicting the host of origin of isolates and outperformed nearest-neighbour phylogenetic host prediction as well as models based on SNPs or AMR data. However, the models did not yield reliable predictions when tested with isolates that were phylogenetically distinct from the training set. The IGR and PV models were often able to differentiate human isolates in clusters where the majority of isolates were from a single animal source. Notably, IGRs were the feature with the best performance across multiple models which may be due to IGRs acting as both a representation of their flanking genes, equivalent to PVs, while also capturing genomic regulatory variation, such as altered promoter regions. The IGR and PV models predict that ~45 % of the human infections with STm in the USA originate from bovine, ~40 % from poultry and ~14.5 % from swine, although sequences of isolates from other sources were not used for training. In summary, the research demonstrates a significant gain in accuracy for models with IGRs and PVs as features compared to SNP-based and core genome phylogeny predictions when applied within the existing population structure. This article contains data hosted by Microreact.

Comparison of Source Attribution Methodologies for Human Campylobacteriosis

Campylobacter spp. are the most common cause of bacterial gastrointestinal infection in humans both in Denmark and worldwide. Studies have found microbial subtyping to be a powerful tool for source attribution, but comparisons of different methodologies are limited. In this study, we compare three source attribution approaches (Machine Learning, Network Analysis, and Bayesian modeling) using three types of whole genome sequences (WGS) data inputs (cgMLST, 5-Mers and 7-Mers). We predicted and compared the sources of human campylobacteriosis cases in Denmark. Using 7mer as an input feature provided the best model performance. The network analysis algorithm had a CSC value of 78.99% and an F1-score value of 67%, while the machine-learning algorithm showed the highest accuracy (98%). The models attributed between 965 and all of the 1224 human cases to a source (network applying 5mer and machine learning applying 7mer, respectively). Chicken from Denmark was the primary source of human campylobacteriosis with an average percentage probability of attribution of 45.8% to 65.4%, representing Bayesian with 7mer and machine learning with cgMLST, respectively. Our results indicate that the different source attribution methodologies based on WGS have great potential for the surveillance and source tracking of Campylobacter. The results of such models may support decision makers to prioritize and target interventions.

Reduced immunogenicity of a live Salmonella enterica serovar Typhimurium vaccine in aged mice

Introduction

Non-typhoidal Salmonella (NTS) is responsible for a high burden of foodborne infections and deaths worldwide. In the United States, NTS infections are the leading cause of hospitalizations and deaths due to foodborne illnesses, and older adults (≥65 years) are disproportionately affected by Salmonella infections. Due to this public health concern, we have developed a live attenuated vaccine, CVD 1926 (I77 ΔguaBA ΔclpP ΔpipA ΔhtrA), against Salmonella enterica serovar Typhimurium, a common serovar of NTS. Little is known about the effect of age on oral vaccine responses, and due to the decline in immune function with age, it is critical to evaluate vaccine candidates in older age groups during early product development.

Methods

In this study, adult (six-to-eight-week-old) and aged (18-month-old) C57BL/6 mice received two doses of CVD 1926 (109 CFU/dose) or PBS perorally, and animals were evaluated for antibody and cell-mediated immune responses. A separate set of mice were immunized and then pre-treated with streptomycin and challenged orally with 108 CFU of wild-type S. Typhimurium SL1344 at 4 weeks postimmunization.

Results

Compared to PBS-immunized mice, adult mice immunized with CVD 1926 had significantly lower S. Typhimurium counts in the spleen, liver, and small intestine upon challenge. In contrast, there were no differences in bacterial loads in the tissues of vaccinated versus PBS aged mice. Aged mice exhibited reduced Salmonella-specific antibody titers in the serum and feces following immunization with CVD 1926 compared to adult mice. In terms of T cell responses (T-CMI), immunized adult mice showed an increase in the frequency of IFN-γ- and IL-2-producing splenic CD4 T cells, IFN-γ- and TNF-α-producing Peyer's Patch (PP)-derived CD4 T cells, and IFN-γ- and TNF-α-producing splenic CD8 T cells compared to adult mice administered PBS. In contrast, in aged mice, T-CMI responses were similar in vaccinated versus PBS mice. CVD 1926 elicited significantly more PP-derived multifunctional T cells in adult compared to aged mice.

Conclusion

These data suggest that our candidate live attenuated S. Typhimurium vaccine, CVD 1926, may not be sufficiently protective or immunogenic in older humans and that mucosal responses to live-attenuated vaccines decrease with increasing age.

Increased phage resistance through lysogenic conversion accompanying emergence of monophasic Salmonella Typhimurium ST34 pandemic strain

Salmonella enterica serovar Typhimurium (S. Typhimurium) comprises a group of closely related human and animal pathogens that account for a large proportion of all Salmonella infections globally. The epidemiological record of S. Typhimurium in Europe is characterized by successive waves of dominant clones, each prevailing for approximately 10-15 years before replacement. Succession of epidemic clones may represent a moving target for interventions aimed at controlling the spread and impact of this pathogen on human and animal health. Here, we investigate the relationship of phage sensitivity and population structure of S. Typhimurium using data from the Anderson phage typing scheme. We observed greater resistance to phage predation of epidemic clones circulating in livestock over the past decades compared to variants with a restricted host range implicating increased resistance to phage in the emergence of epidemic clones of particular importance to human health. Emergence of monophasic S. Typhimurium ST34, the most recent dominant multidrug-resistant clone, was accompanied by increased resistance to phage predation during clonal expansion, in part by the acquisition of the mTmII prophage that may have contributed to the fitness of the strains that replaced ancestors lacking this prophage.

WGS-Based Lineage and Antimicrobial Resistance Pattern of Salmonella Typhimurium Isolated during 2000-2017 in Peru

Salmonella Typhimurium is associated with foodborne diseases worldwide, including in Peru, and its emerging antibiotic resistance (AMR) is now a global public health problem. Therefore, country-specific monitoring of the AMR emergence is vital to control this pathogen, and in these aspects, whole genome sequence (WGS)—based approaches are better than gene-based analyses. Here, we performed the antimicrobial susceptibility test for ten widely used antibiotics and WGS-based various analyses of 90 S. Typhimurium isolates (human, animal, and environment) from 14 cities of Peru isolated from 2000 to 2017 to understand the lineage and antimicrobial resistance pattern of this pathogen in Peru. Our results suggest that the Peruvian isolates are of Typhimurium serovar and predominantly belong to sequence type ST19. Genomic diversity analyses indicate an open pan-genome, and at least ten lineages are circulating in Peru. A total of 48.8% and 31.0% of isolates are phenotypically and genotypically resistant to at least one antibiotic, while 12.0% are multi-drug resistant (MDR). Genotype−phenotype correlations for ten tested drugs show >80% accuracy, and >90% specificity. Sensitivity above 90% was only achieved for ciprofloxacin and ceftazidime. Two lineages exhibit the majority of the MDR isolates. A total of 63 different AMR genes are detected, of which 30 are found in 17 different plasmids. Transmissible plasmids such as lncI-gamma/k, IncI1-I(Alpha), Col(pHAD28), IncFIB, IncHI2, and lncI2 that carry AMR genes associated with third-generation antibiotics are also identified. Finally, three new non-synonymous single nucleotide variations (SNVs) for nalidixic acid and eight new SNVs for nitrofurantoin resistance are predicted using genome-wide association studies, comparative genomics, and functional annotation. Our analysis provides for the first time the WGS-based details of the circulating S. Typhimurium lineages and their antimicrobial resistance pattern in Peru.

Emergence, Dissemination and Antimicrobial Resistance of the Main Poultry-Associated Salmonella Serovars in Brazil

Simple Summary

Salmonellosis is a human and animal disease caused by Salmonella, a bacterial genus classified into different species, subspecies, and serological variants (serovars) according to adaptation to one or more different hosts (animals and humans), pathogenicity profiles, and antigenic properties. Some specific Salmonella serovars can spread more easily in the enteric microbiota of avian species, often causing disease in birds and/or being transmitted to humans through food (such as chicken and eggs). Antimicrobial resistance (AMR) has also been reported in poultry-associated Salmonella isolates due to the widespread use of antimicrobials on farms. The availability of comprehensive data on the emergence and spread of Salmonella serovars, as well as their AMR profiles in farms and food products in Brazil (a major producer of poultry in the World), is necessary to understand their relevance in all avian production chains and also occurrence in poultry-derived foods. This article aims to provide an overview of the genus Salmonella and the main serovars that emerged in Brazilian poultry over time (Gallinarum, Typhimurium, Enteritidis, Heidelberg, and Minnesota), reviewing the scientific literature and suggesting more effective prevention and control for the future.

Abstract

Salmonella infects poultry, and it is also a human foodborne pathogen. This bacterial genus is classified into several serovars/lineages, some of them showing high antimicrobial resistance (AMR). The ease of Salmonella transmission in farms, slaughterhouses, and eggs industries has made controlling it a real challenge in the poultry-production chains. This review describes the emergence, dissemination, and AMR of the main Salmonella serovars and lineages detected in Brazilian poultry. It is reported that few serovars emerged and have been more widely disseminated in breeders, broilers, and layers in the last 70 years. Salmonella Gallinarum was the first to spread on the farms, remaining as a concerning poultry pathogen. Salmonella Typhimurium and Enteritidis were also largely detected in poultry and foods (eggs, chicken, turkey), being associated with several human foodborne outbreaks. Salmonella Heidelberg and Minnesota have been more widely spread in recent years, resulting in frequent chicken/turkey meat contamination. A few more serovars (Infantis, Newport, Hadar, Senftenberg, Schwarzengrund, and Mbandaka, among others) were also detected, but less frequently and usually in specific poultry-production regions. AMR has been identified in most isolates, highlighting multi-drug resistance in specific poultry lineages from the serovars Typhimurium, Heidelberg, and Minnesota. Epidemiological studies are necessary to trace and control this pathogen in Brazilian commercial poultry production chains.

Comparative Genomic Analysis Discloses Differential Distribution of Antibiotic Resistance Determinants between Worldwide Strains of the Emergent ST213 Genotype of Salmonella Typhimurium

Salmonella enterica constitutes a global public health concern as one of the main etiological agents of human gastroenteritis. The Typhimurium serotype is frequently isolated from human, animal, food, and environmental samples, with its sequence type 19 (ST19) being the most widely distributed around the world as well as the founder genotype. The replacement of the ST19 genotype with the ST213 genotype that has multiple antibiotic resistance (MAR) in human and food samples was first observed in Mexico. The number of available genomes of ST213 strains in public databases indicates its fast worldwide dispersion, but its public health relevance is unknown. A comparative genomic analysis conducted as part of this research identified the presence of 44 genes, 34 plasmids, and five point mutations associated with antibiotic resistance, distributed across 220 genomes of ST213 strains, indicating the MAR phenotype. In general, the grouping pattern in correspondence to the presence/absence of genes/plasmids that confer antibiotic resistance cluster the genomes according to the geographical origin where the strain was isolated. Genetic determinants of antibiotic resistance group the genomes of North America (Canada, Mexico, USA) strains, and suggest a dispersion route to reach the United Kingdom and, from there, the rest of Europe, then Asia and Oceania. The results obtained here highlight the worldwide public health relevance of the ST213 genotype, which contains a great diversity of genetic elements associated with MAR.

The Number and Type of Chaperone-Usher Fimbriae Reflect Phylogenetic Clade Rather than Host Range in Salmonella

Salmonella is one of the most successful foodborne pathogens worldwide, owing in part to its ability to colonize or infect a wide range of hosts. Salmonella serovars are known to encode a variety of different fimbriae (hairlike organelles that facilitate binding to surfaces); however, the distribution, number, and sequence diversity of fimbriae encoded across different lineages of Salmonella were unknown. We queried whole-genome sequence (WGS) data for 242 Salmonella enterica subsp. enterica (subspecies enterica) isolates from the top 217 serovars associated with isolation from humans and agricultural animals; this effort identified 2,894 chaperone-usher (CU)-type fimbrial usher sequences, representing the most conserved component of CU fimbriae. On average, isolates encoded 12 different CU fimbrial ushers (6 to 18 per genome), although the distribution varied significantly (P = 1.328E-08) by phylogenetic clade, with isolates in section Typhi having significantly fewer fimbrial ushers than isolates in clade A2 (medians = 10 and 12 ushers, respectively). Characterization of fimbriae in additional non-enterica subspecies genomes suggested that 8 fimbrial ushers were classified as being unique to subspecies enterica isolates, suggesting that the majority of fimbriae were most likely acquired prior to the divergence of subspecies enterica. Characterization of mobile elements suggested that plasmids represent an important vehicle facilitating the acquisition of a wide range of fimbrial ushers, particularly for the acquisition of fimbriae from other Gram-negative genera. Overall, our results suggest that differences in the number and type of fimbriae encoded most likely reflect differences in phylogenetic clade rather than differences in host range. IMPORTANCE Fimbriae of the CU assembly pathway represent important organelles that mediate Salmonella's interactions with host tissues and abiotic surfaces. Our analyses provide a comprehensive overview of the diversity of CU fimbriae in Salmonella spp., highlighting that the majority of CU fimbriae are distributed broadly across multiple subspecies and suggesting that acquisition most likely occurred prior to the divergence of subspecies enterica. Our data also suggest that plasmids represent the primary vehicles facilitating the horizontal transfer of diverse CU fimbriae in Salmonella. Finally, the observed high sequence similarity between some ushers suggests that different names may have been assigned to closely related fimbrial ushers that likely should be represented by a single designation. This highlights the need to establish standard criteria for fimbria classification and nomenclature, which will also facilitate future studies seeking to associate virulence factors with adaptation to or differences in the likelihood of causing disease in a given host.

A Machine Learning Model for Food Source Attribution of Listeria monocytogenes

Despite its low morbidity, listeriosis has a high mortality rate due to the severity of its clinical manifestations. The source of human listeriosis is often unclear. In this study, we investigate the ability of machine learning to predict the food source from which clinical Listeria monocytogenes isolates originated. Four machine learning classification algorithms were trained on core genome multilocus sequence typing data of 1212 L. monocytogenes isolates from various food sources. The average accuracies of random forest, support vector machine radial kernel, stochastic gradient boosting, and logit boost were found to be 0.72, 0.61, 0.7, and 0.73, respectively. Logit boost showed the best performance and was used in model testing on 154 L. monocytogenes clinical isolates. The model attributed 17.5 % of human clinical cases to dairy, 32.5% to fruits, 14.3% to leafy greens, 9.7% to meat, 4.6% to poultry, and 18.8% to vegetables. The final model also provided us with genetic features that were predictive of specific sources. Thus, this combination of genomic data and machine learning-based models can greatly enhance our ability to track L. monocytogenes from different food sources.

Source Attribution of Human Campylobacteriosis Using Whole-Genome Sequencing Data and Network Analysis

Campylobacter spp. are a leading and increasing cause of gastrointestinal infections worldwide. Source attribution, which apportions human infection cases to different animal species and food reservoirs, has been instrumental in control- and evidence-based intervention efforts. The rapid increase in whole-genome sequencing data provides an opportunity for higher-resolution source attribution models. Important challenges, including the high dimension and complex structure of WGS data, have inspired concerted research efforts to develop new models. We propose network analysis models as an accurate, high-resolution source attribution approach for the sources of human campylobacteriosis. A weighted network analysis approach was used in this study for source attribution comparing different WGS data inputs. The compared model inputs consisted of cgMLST and wgMLST distance matrices from 717 human and 717 animal isolates from cattle, chickens, dogs, ducks, pigs and turkeys. SNP distance matrices from 720 human and 720 animal isolates were also used. The data were collected from 2015 to 2017 in Denmark, with the animal sources consisting of domestic and imports from 7 European countries. Clusters consisted of network nodes representing respective genomes and links representing distances between genomes. Based on the results, animal sources were the main driving factor for cluster formation, followed by type of species and sampling year. The coherence source clustering (CSC) values based on animal sources were 78%, 81% and 78% for cgMLST, wgMLST and SNP, respectively. The CSC values based on Campylobacter species were 78%, 79% and 69% for cgMLST, wgMLST and SNP, respectively. Including human isolates in the network resulted in 88%, 77% and 88% of the total human isolates being clustered with the different animal sources for cgMLST, wgMLST and SNP, respectively. Between 12% and 23% of human isolates were not attributed to any animal source. Most of the human genomes were attributed to chickens from Denmark, with an average attribution percentage of 52.8%, 52.2% and 51.2% for cgMLST, wgMLST and SNP distance matrices respectively, while ducks from Denmark showed the least attribution of 0% for all three distance matrices. The best-performing model was the one using wgMLST distance matrix as input data, which had a CSC value of 81%. Results from our study show that the weighted network-based approach for source attribution is reliable and can be used as an alternative method for source attribution considering the high performance of the model. The model is also robust across the different Campylobacter species, animal sources and WGS data types used as input.

Metagenomics and artificial intelligence in the context of human health

Human microbiome is ubiquitous, dynamic, and site-specific consortia of microbial communities. The pathogenic nature of microorganisms within human tissues has led to an increase in microbial studies. Characterization of genera, like Streptococcus, Cutibacterium, Staphylococcus, Bifidobacterium, Lactococcus and Lactobacillus through culture-dependent and culture-independent techniques has been reported. However, due to the unique environment within human tissues, it is difficult to culture these microorganisms making their molecular studies strenuous. MGs offer a gateway to explore and characterize hidden microbial communities through a culture-independent mode by direct DNA isolation. By function and sequence-based MGs, Scientists can explore the mechanistic details of numerous microbes and their interaction with the niche. Since the data generated from MGs studies is highly complex and multi-dimensional, it requires accurate analytical tools to evaluate and interpret the data. Artificial intelligence (AI) provides the luxury to automatically learn the data dimensionality and ease its complexity that makes the disease diagnosis and disease response easy, accurate and timely. This review provides insight into the human microbiota and its exploration and expansion through MG studies. The review elucidates the significance of MGs in studying the changing microbiota during disease conditions besides highlighting the role of AI in computational analysis of MG data.

Prevalence and Genomic Investigation of Multidrug-Resistant Salmonella Isolates from Companion Animals in Hangzhou, China

Salmonella is a group of bacteria that constitutes the leading cause of diarrheal diseases, posing a great disease burden worldwide. There are numerous pathways for zoonotic Salmonella transmission to humans; however, the role of companion animals in spreading these bacteria is largely underestimated in China. We aimed to investigate the prevalence of Salmonella in pet dogs and cats in Hangzhou, China, and characterize the antimicrobial resistance profile and genetic features of these pet-derived pathogens. In total, 137 fecal samples of pets were collected from an animal hospital in Hangzhou in 2018. The prevalence of Salmonella was 5.8% (8/137) in pets, with 9.3% (5/54) of cats and 3.6% (3/83) of dogs being Salmonella positive. By whole-genome sequencing (WGS), in silico serotyping, and multilocus sequence typing (MLST), 26 pet-derived Salmonella isolates were identified as Salmonella Dublin (ST10, n = 22) and Salmonella Typhimurium (ST19, n = 4). All of the isolates were identified as being multidrug-resistant (MDR), by conducting antimicrobial susceptibility testing under both aerobic and anaerobic conditions. The antibiotics of the most prevalent resistance were streptomycin (100%), cotrimoxazole (100%), tetracycline (96.20%), and ceftriaxone (92.30%). Versatile antimicrobial-resistant genes were identified, including floR (phenicol-resistant gene), blaCTX-M-15, and blaCTX-M-55 (extended-spectrum beta-lactamase genes). A total of 11 incompatible (Inc) plasmids were identified, with IncA/C2, IncFII(S), and IncX1 being the most predominant among Salmonella Dublin, and IncFIB(S), IncFII(S), IncI1, and IncQ1 being the most prevailing among Salmonella Typhimurium. Our study applied WGS to characterize pet-derived Salmonella in China, showing the presence of MDR Salmonella in pet dogs and cats with a high diversity of ARGs and plasmids. These data indicate a necessity for the regular surveillance of pet-derived pathogens to mitigate zoonotic diseases.

Use of Whole Genome Sequencing by the Federal Interagency Collaboration for Genomics for Food and Feed Safety in the United States

ABSTRACT

This multiagency report developed by the Interagency Collaboration for Genomics for Food and Feed Safety provides an overview of the use of and transition to whole genome sequencing (WGS) technology for detection and characterization of pathogens transmitted commonly by food and for identification of their sources. We describe foodborne pathogen analysis, investigation, and harmonization efforts among the following federal agencies: National Institutes of Health; Department of Health and Human Services, Centers for Disease Control and Prevention (CDC) and U.S. Food and Drug Administration (FDA); and the U.S. Department of Agriculture, Food Safety and Inspection Service, Agricultural Research Service, and Animal and Plant Health Inspection Service. We describe single nucleotide polymorphism, core-genome, and whole genome multilocus sequence typing data analysis methods as used in the PulseNet (CDC) and GenomeTrakr (FDA) networks, underscoring the complementary nature of the results for linking genetically related foodborne pathogens during outbreak investigations while allowing flexibility to meet the specific needs of Interagency Collaboration partners. We highlight how we apply WGS to pathogen characterization (virulence and antimicrobial resistance profiles) and source attribution efforts and increase transparency by making the sequences and other data publicly available through the National Center for Biotechnology Information. We also highlight the impact of current trends in the use of culture-independent diagnostic tests for human diagnostic testing on analytical approaches related to food safety and what is next for the use of WGS in the area of food safety.

HIGHLIGHTS

Impact of long-term dietary habits on the human gut resistome in the Dutch population

The human gut microbiome plays a central role in health and disease. Environmental factors, such as lifestyle and diet, are known to shape the gut microbiome as well as the reservoir of resistance genes that these microbes harbour; the resistome. In this study we assessed whether long-term dietary habits within a single geographical region (the Netherlands) impact the human gut resistome. Faecal samples from Dutch omnivores, pescatarians, vegetarians and vegans were analysed by metagenomic shotgun sequencing (MSS) (n = 149) and resistome capture sequencing approach (ResCap) (n = 64). Among all diet groups, 119 and 145 unique antibiotic resistance genes (ARGs) were detected by MSS or ResCap, respectively. Five or fifteen ARGs were shared between all diet groups, based on MSS and ResCap, respectively. The total number of detected ARGs by MSS or ResCap was not significantly different between the groups. MSS also revealed that vegans have a distinct microbiome composition, compared to other diet groups. Vegans had a lower abundance of Streptococcus thermophilus and Lactococcus lactis compared to pescatarians and a lower abundance of S. thermophilus when compared to omnivores. In summary, our study showed that long-term dietary habits are not associated with a specific resistome signature.

Salmonella enterica serovar Typhimurium from Wild Birds in the United States Represent Distinct Lineages Defined by Bird Type

Salmonella enterica serovar Typhimurium is typically considered a host generalist; however, certain isolates are associated with specific hosts and show genetic features of host adaptation. Here, we sequenced 131 S. Typhimurium isolates from wild birds collected in 30 U.S. states during 1978–2019. We found that isolates from broad taxonomic host groups including passerine birds, water birds (Aequornithes), and larids (gulls and terns) represented three distinct lineages and certain S. Typhimurium CRISPR types presented in individual lineages. We also showed that lineages formed by wild bird isolates differed from most isolates originating from domestic animal sources, and that genomes from these lineages substantially improved source attribution of Typhimurium genomes to wild birds by a machine learning classifier. Furthermore, virulence gene signatures that differentiated S. Typhimurium from passerines, water birds, and larids were detected. Passerine isolates tended to lack S. Typhimurium-specific virulence plasmids. Isolates from the passerine, water bird, and larid lineages had close genetic relatedness with human clinical isolates, including those from a 2021 U.S. outbreak linked to passerine birds. These observations indicate that S. Typhimurium from wild birds in the United States are likely host-adapted, and the representative genomic data set examined in this study can improve source prediction and facilitate outbreak investigation.

IMPORTANCE Within-host evolution of S. Typhimurium may lead to pathovars adapted to specific hosts. Here, we report the emergence of disparate avian S. Typhimurium lineages with distinct virulence gene signatures. The findings highlight the importance of wild birds as a reservoir for S. Typhimurium and contribute to our understanding of the genetic diversity of S. Typhimurium from wild birds. Our study indicates that S. Typhimurium may have undergone adaptive evolution within wild birds in the United States. The representative S. Typhimurium genomes from wild birds, together with the virulence gene signatures identified in these bird isolates, are valuable for S. Typhimurium source attribution and epidemiological surveillance.

Salmonella Genomics in Public Health and Food Safety

The species Salmonella enterica comprises over 2,600 serovars, many of which are known to be intracellular pathogens of mammals, birds, and reptiles. It is now apparent that Salmonella is a highly adapted environmental microbe and can readily persist in a number of environmental niches, including water, soil, and various plant (including produce) species. Much of what is known about the evolution and diversity of nontyphoidal Salmonella serovars (NTS) in the environment is the result of the rise of the genomics era in enteric microbiology. There are over 340,000 Salmonella genomes available in public databases. This extraordinary breadth of genomic diversity now available for the species, coupled with widespread availability and affordability of whole-genome sequencing (WGS) instrumentation, has transformed the way in which we detect, differentiate, and characterize Salmonella enterica strains in a timely way. Not only have WGS data afforded a detailed and global examination of the molecular epidemiological movement of Salmonella from diverse environmental reservoirs into human and animal hosts, but they have also allowed considerable consolidation of the diagnostic effort required to test for various phenotypes important to the characterization of Salmonella. For example, drug resistance, serovar, virulence determinants, and other genome-based attributes can all be discerned using a genome sequence. Finally, genomic analysis, in conjunction with functional and phenotypic approaches, is beginning to provide new insights into the precise adaptive changes that permit persistence of NTS in so many diverse and challenging environmental niches.

Application of machine learning to the monitoring and prediction of food safety: A review

Machine learning (ML) has proven to be a useful technology for data analysis and modeling in a wide variety of domains, including food science and engineering. The use of ML models for the monitoring and prediction of food safety is growing in recent years. Currently, several studies have reviewed ML applications on foodborne disease and deep learning applications on food. This article presents a literature review on ML applications for monitoring and predicting food safety. The paper summarizes and categorizes ML applications in this domain, categorizes and discusses data types used for ML modeling, and provides suggestions for data sources and input variables for future ML applications. The review is based on three scientific literature databases: Scopus, CAB Abstracts, and IEEE. It includes studies that were published in English in the period from January 1, 2011 to April 1, 2021. Results show that most studies applied Bayesian networks, Neural networks, or Support vector machines. Of the various ML models reviewed, all relevant studies showed high prediction accuracy by the validation process. Based on the ML applications, this article identifies several avenues for future studies applying ML models for the monitoring and prediction of food safety, in addition to providing suggestions for data sources and input variables.

A genomic data resource for predicting antimicrobial resistance from laboratory-derived antimicrobial susceptibility phenotypes

Antimicrobial resistance (AMR) is a major global health threat that affects millions of people each year. Funding agencies worldwide and the global research community have expended considerable capital and effort tracking the evolution and spread of AMR by isolating and sequencing bacterial strains and performing antimicrobial susceptibility testing (AST). For the last several years, we have been capturing these efforts by curating data from the literature and data resources and building a set of assembled bacterial genome sequences that are paired with laboratory-derived AST data. This collection currently contains AST data for over 67 000 genomes encompassing approximately 40 genera and over 100 species. In this paper, we describe the characteristics of this collection, highlighting areas where sampling is comparatively deep or shallow, and showing areas where attention is needed from the research community to improve sampling and tracking efforts. In addition to using the data to track the evolution and spread of AMR, it also serves as a useful starting point for building machine learning models for predicting AMR phenotypes. We demonstrate this by describing two machine learning models that are built from the entire dataset to show where the predictive power is comparatively high or low. This AMR metadata collection is freely available and maintained on the Bacterial and Viral Bioinformatics Center (BV-BRC) FTP site ftp://ftp.bvbrc.org/RELEASE_NOTES/PATRIC_genomes_AMR.txt.

Genomic signatures of host adaptation in group B Salmonella enterica ST416/ST417 from harbour porpoises

A type of monophasic group B Salmonella enterica with the antigenic formula 4,12:a:- (“Fulica-like”) has been described as associated with harbour porpoises (Phocoena phocoena), most frequently recovered from lung samples. In the present study, lung tissue samples from 47 porpoises found along the Swedish coast or as bycatch in fishing nets were analysed, two of which were positive for S. enterica. Pneumonia due to the infection was considered the likely cause of death for one of the two animals. The recovered isolates were whole genome sequenced and found to belong to sequence type (ST) 416 and to be closely related to ST416/ST417 porpoise isolates from UK waters as determined by core-genome MLST. Serovars Bispebjerg, Fulica and Abortusequi were identified as distantly related to the porpoise isolates, but no close relatives from other host species were found. All ST416/417 isolates had extensive loss of function mutations in key Salmonella pathogenicity islands, but carried accessory genetic elements associated with extraintestinal infection such as iron uptake systems. Gene ontology and pathway analysis revealed reduced secondary metabolic capabilities and loss of function in terms of signalling and response to environmental cues, consistent with adaptation for the extraintestinal niche. A classification system based on machine learning identified ST416/417 as more invasive than classical gastrointestinal serovars. Genome analysis results are thus consistent with ST416/417 as a host-adapted and extraintestinal clonal population of S. enterica, which while found in porpoises without associated pathology can also cause severe opportunistic infections.

Machine learning to predict the source of campylobacteriosis using whole genome data

Campylobacteriosis is among the world's most common foodborne illnesses, caused predominantly by the bacterium Campylobacter jejuni. Effective interventions require determination of the infection source which is challenging as transmission occurs via multiple sources such as contaminated meat, poultry, and drinking water. Strain variation has allowed source tracking based upon allelic variation in multi-locus sequence typing (MLST) genes allowing isolates from infected individuals to be attributed to specific animal or environmental reservoirs. However, the accuracy of probabilistic attribution models has been limited by the ability to differentiate isolates based upon just 7 MLST genes. Here, we broaden the input data spectrum to include core genome MLST (cgMLST) and whole genome sequences (WGS), and implement multiple machine learning algorithms, allowing more accurate source attribution. We increase attribution accuracy from 64% using the standard iSource population genetic approach to 71% for MLST, 85% for cgMLST and 78% for kmerized WGS data using the classifier we named aiSource. To gain insight beyond the source model prediction, we use Bayesian inference to analyse the relative affinity of C. jejuni strains to infect humans and identified potential differences, in source-human transmission ability among clonally related isolates in the most common disease causing lineage (ST-21 clonal complex). Providing generalizable computationally efficient methods, based upon machine learning and population genetics, we provide a scalable approach to global disease surveillance that can continuously incorporate novel samples for source attribution and identify fine-scale variation in transmission potential.

Global spread of Salmonella Enteritidis via centralized sourcing and international trade of poultry breeding stocks

A pandemic of Salmonella enterica serotype Enteritidis emerged in the 1980s due to contaminated poultry products. How Salmonella Enteritidis rapidly swept through continents remains a historical puzzle as the pathogen continues to cause outbreaks and poultry supply becomes globalized. We hypothesize that international trade of infected breeding stocks causes global spread of the pathogen. By integrating over 30,000 Salmonella Enteritidis genomes from 98 countries during 1949-2020 and international trade of live poultry from the 1980s to the late 2010s, we present multifaceted evidence that converges on a high likelihood, global scale, and extended protraction of Salmonella Enteritidis dissemination via centralized sourcing and international trade of breeding stocks. We discovered recent, genetically near-identical isolates from domestically raised poultry in North and South America. We obtained phylodynamic characteristics of global Salmonella Enteritidis populations that lend spatiotemporal support for its dispersal from centralized origins during the pandemic. We identified concordant patterns of international trade of breeding stocks and quantitatively established a driving role of the trade in the geographic dispersal of Salmonella Enteritidis, suggesting that the centralized origins were infected breeding stocks. Here we demonstrate the value of integrative and hypothesis-driven data mining in unravelling otherwise difficult-to-probe pathogen dissemination from hidden origins.

Elucidation of global and national genomic epidemiology of Salmonella enterica serovar Enteritidis through multilevel genome typing

Salmonella enterica serovar Enteritidis is a major cause of foodborne Salmonella infections and outbreaks in humans. Effective surveillance and timely outbreak detection are essential for public health control. Multilevel genome typing (MGT) with multiple levels of resolution has been previously demonstrated as a promising tool for this purpose. In this study, we developed MGT with nine levels for S. Enteritidis and characterised the genomic epidemiology of S. Enteritidis in detail. We examined 26 670 publicly available S. Enteritidis genome sequences from isolates spanning 101 years from 86 countries to reveal their spatial and temporal distributions. Using the lower resolution MGT levels, globally prevalent and regionally restricted sequence types (STs) were identified; avian associated MGT4-STs were found that were common in human cases in the USA; temporal trends were observed in the UK with MGT5-STs from 2014 to 2018 revealing both long lived endemic STs and the rapid expansion of new STs. Using MGT3 to MGT6, we identified multidrug resistance (MDR) associated STs at various MGT levels, which improves precision of detection and global tracking of MDR clones. We also found that the majority of the global S. Enteritidis population fell within two predominant lineages, which had significantly different propensity of causing large scale outbreaks. An online open MGT database has been established for unified international surveillance of S. Enteritidis. We demonstrated that MGT provides a flexible and high-resolution genome typing tool for S. Enteritidis surveillance and outbreak detection.

Salmonella pathogenesis and host-adaptation in farmed animals

Salmonella is an animal and zoonotic pathogen of global importance. Depending on pathogen and host factors, infections can be asymptomatic or involve acute gastroenteritis or invasive disease. Genomic signatures associated with host-range, tissue tropism or differential virulence of Salmonella enterica serovars, and their variants, have emerged. In turn, it is becoming feasible to predict invasive potential, host-adaptation and zoonotic risk of Salmonella from sequence data to improve outbreak investigation, risk assessment and control strategies. Functional annotation of Salmonella genomes has accelerated with the screening of high-density mutant libraries, revealing host-specific, niche-specific and serovar-specific virulence factors. As natural hosts and reservoirs, farmed animals provide powerful insights into host-adaptation and pathogenesis of Salmonella not always evident from surrogate rodent or cell-based models.

Enhancing genomics-based outbreak detection of endemic Salmonella enterica serovar Typhimurium using dynamic thresholds

Salmonella enterica serovar Typhimurium is the leading cause of salmonellosis in Australia, and the ability to identify outbreaks and their sources is vital to public health. Here, we examined the utility of whole-genome sequencing (WGS), including complete genome sequencing with Oxford Nanopore technologies, in examining 105 isolates from an endemic multi-locus variable number tandem repeat analysis (MLVA) type over 5 years. The MLVA type was very homogeneous, with 90 % of the isolates falling into groups with a five SNP cut-off. We developed a new two-step approach for outbreak detection using WGS. The first clustering at a zero single nucleotide polymorphism (SNP) cut-off was used to detect outbreak clusters that each occurred within a 4 week window and then a second clustering with dynamically increased SNP cut-offs were used to generate outbreak investigation clusters capable of identifying all outbreak cases. This approach offered optimal specificity and sensitivity for outbreak detection and investigation, in particular of those caused by endemic MLVA types or clones with low genetic diversity. We further showed that inclusion of complete genome sequences detected no additional mutational events for genomic outbreak surveillance. Phylogenetic analysis found that the MLVA type was likely to have been derived recently from a single source that persisted over 5 years, and seeded numerous sporadic infections and outbreaks. Our findings suggest that SNP cut-offs for outbreak cluster detection and public-health surveillance should be based on the local diversity of the relevant strains over time. These findings have general applicability to outbreak detection of bacterial pathogens.

Interpretability Versus Accuracy: A Comparison of Machine Learning Models Built Using Different Algorithms, Performance Measures, and Features to Predict E. coli Levels in Agricultural Water

Since E. coli is considered a fecal indicator in surface water, government water quality standards and industry guidance often rely on E. coli monitoring to identify when there is an increased risk of pathogen contamination of water used for produce production (e.g., for irrigation). However, studies have indicated that E. coli testing can present an economic burden to growers and that time lags between sampling and obtaining results may reduce the utility of these data. Models that predict E. coli levels in agricultural water may provide a mechanism for overcoming these obstacles. Thus, this proof-of-concept study uses previously published datasets to train, test, and compare E. coli predictive models using multiple algorithms and performance measures. Since the collection of different feature data carries specific costs for growers, predictive performance was compared for models built using different feature types [geospatial, water quality, stream traits, and/or weather features]. Model performance was assessed against baseline regression models. Model performance varied considerably with root-mean-squared errors and Kendall's Tau ranging between 0.37 and 1.03, and 0.07 and 0.55, respectively. Overall, models that included turbidity, rain, and temperature outperformed all other models regardless of the algorithm used. Turbidity and weather factors were also found to drive model accuracy even when other feature types were included in the model. These findings confirm previous conclusions that machine learning models may be useful for predicting when, where, and at what level E. coli (and associated hazards) are likely to be present in preharvest agricultural water sources. This study also identifies specific algorithm-predictor combinations that should be the foci of future efforts to develop deployable models (i.e., models that can be used to guide on-farm decision-making and risk mitigation). When deploying E. coli predictive models in the field, it is important to note that past research indicates an inconsistent relationship between E. coli levels and foodborne pathogen presence. Thus, models that predict E. coli levels in agricultural water may be useful for assessing fecal contamination status and ensuring compliance with regulations but should not be used to assess the risk that specific pathogens of concern (e.g., Salmonella, Listeria) are present.

Detection of spatial and spatio-temporal Salmonella Heidelberg and Salmonella Typhimurium human case clusters focused around licensed abattoirs in Ontario in 2015, and their potential relation to known outbreaks

Salmonellosis is one of several zoonotic diseases for which individuals with occupational animal contact, including abattoir workers, are at an increased risk. If meat is contaminated during slaughter, this can increase the risk of enteric illness for consumers. In this study, we investigated whether reported cases of Salmonella Heidelberg and Typhimurium were clustered around abattoirs in Ontario in 2015 and whether there was any evidence (laboratory/exposure) to suggest an abattoir at the centre of a cluster might be the source of exposure. Data for each reported case of S. Heidelberg and S. Typhimurium in Ontario in 2015 were collected. Multi-focused and non-focused spatial and space-time cluster detection tests were performed for each serotype, with and without cases linked to known outbreaks, using Poisson and space-time permutation models. Focused tests included the location of abattoirs operational in all or part of 2015. Laboratory data and exposure information were used to explore the relatedness of cases within identified clusters. Focused spatial tests identified clusters of S. Heidelberg and S. Typhimurium around abattoirs. Focused space-time permutation tests identified 2 significant space-time clusters of S. Heidelberg; one cluster (n = 11 cases) included 8 of 9 cases associated with a known outbreak and the other cluster (n = 18 cases) was not part of a previously identified outbreak. Review of laboratory and risk factor information suggested that cases within each cluster shared a common exposure. Cases were not asked about goat or sheep meat consumption. The focused cluster test, particularly with the space-time permutation model, could assist in identifying outbreaks associated with a particular physical location, such as an abattoir. Improvements to the current case investigation process, such as consistent collection and reporting of high-risk occupation information and more detailed food consumption history, could assist in outbreak identification when coupled with this statistic.

Ecological niche adaptation of Salmonella Typhimurium U288 is associated with altered pathogenicity and reduced zoonotic potential

The emergence of new bacterial pathogens is a continuing challenge for agriculture and food safety. Salmonella Typhimurium is a major cause of foodborne illness worldwide, with pigs a major zoonotic reservoir. Two phylogenetically distinct variants, U288 and ST34, emerged in UK pigs around the same time but present different risk to food safety. Here we show using genomic epidemiology that ST34 accounts for over half of all S. Typhimurium infections in people while U288 less than 2%. That the U288 clade evolved in the recent past by acquiring AMR genes, indels in the virulence plasmid pU288-1, and accumulation of loss-of-function polymorphisms in coding sequences. U288 replicates more slowly and is more sensitive to desiccation than ST34 isolates and exhibited distinct pathogenicity in the murine model of colitis and in pigs. U288 infection was more disseminated in the lymph nodes while ST34 were recovered in greater numbers in the intestinal contents. These data are consistent with the evolution of S. Typhimurium U288 adaptation to pigs that may determine their reduced zoonotic potential.

Salmonella spp. prevalence and antimicrobial resistance in broiler chicken and turkey flocks in Canada from 2013 to 2018

Salmonella infections are a major human health concern. In the elderly and immunocompromised, infections can be life-threatening and may require antibiotic therapy. Where antibiotic therapy is required, antimicrobials of choice include fluoroquinolones and extended-spectrum cephalosporins (ESC). The aim of this study is to utilize data from the Canadian Integrated Program for Antimicrobial Resistance Surveillance (CIPARS) to compare the prevalence of Salmonella serovars between broiler chicken and turkey flocks across Canada, and to gain an understanding of the prevalence of resistance to antimicrobials categorized as important to human health. There were 1,596 Salmonella isolates obtained from 514 broiler chicken flocks, and 659 Salmonella isolates obtained from 217 turkey flocks (2013-2018). All isolates were obtained from pooled faecal samples. Among broiler chicken flocks, the top three serovars were Kentucky (n = 573, 36%), Enteritidis (n = 314, 20%) and Heidelberg (n = 127, 8%). Resistance to ceftriaxone among Salmonella ser. Kentucky decreased from 27% in 2013 to 22% in 2018. There was no resistance among Salmonella ser. Enteritidis reported until 2018 when one isolate from British Columbia was resistant to ampicillin, streptomycin, sulphisoxazole and tetracycline. Salmonella ser. Heidelberg resistance to ceftriaxone decreased from 19% in 2013 to 14% in 2018. Among turkey flocks the top three serovars were Uganda (n = 109, 16.5%), Hadar (n = 85, 12%) and Muenchen (n = 66, 10%). No isolates of Salmonella ser. Uganda or Salmonella ser. Muenchen were resistant to any β-lactams. Salmonella ser. Hadar (34/81, 42%) exhibited resistance to ampicillin. There was no resistance to quinolones among turkey isolates. Emerging resistance among Salmonella ser. Enteritidis, and resistance to β-lactams and fluoroquinolones among Salmonella ser. Kentucky from broilers are cause for concern as these classes of antimicrobials are important for treatment of salmonellosis.

Salmonella Genomics and Population Analyses Reveal High Inter- and Intraserovar Diversity in Freshwater

Freshwater can support the survival of the enteric pathogen Salmonella, though temporal Salmonella diversity in a large watershed has not been assessed. At 28 locations within the Susquehanna River basin, 10-liter samples were assessed in spring and summer over 2 years. Salmonella prevalence was 49%, and increased river discharge was the main driver of Salmonella presence. The amplicon-based sequencing tool, CRISPR-SeroSeq, was used to determine serovar population diversity and detected 25 different Salmonella serovars, including up to 10 serovars from a single water sample. On average, there were three serovars per sample, and 80% of Salmonella-positive samples contained more than one serovar. Serovars Give, Typhimurium, Thompson, and Infantis were identified throughout the watershed and over multiple collections. Seasonal differences were evident: serovar Give was abundant in the spring, whereas serovar Infantis was more frequently identified in the summer. Eight of the ten serovars most commonly associated with human illness were detected in this study. Crucially, six of these serovars often existed in the background, where they were masked by a more abundant serovar(s) in a sample. Serovars Enteritidis and Typhimurium, especially, were masked in 71 and 78% of samples where they were detected, respectively. Whole-genome sequencing-based phylogeny demonstrated that strains within the same serovar collected throughout the watershed were also very diverse. The Susquehanna River basin is the largest system where Salmonella prevalence and serovar diversity have been temporally and spatially investigated, and this study reveals an extraordinary level of inter- and intraserovar diversity.IMPORTANCE Salmonella is a leading cause of bacterial foodborne illness in the United States, and outbreaks linked to fresh produce are increasing. Understanding Salmonella ecology in freshwater is of importance, especially where irrigation practices or recreational use occur. As the third largest river in the United States east of the Mississippi, the Susquehanna River is the largest freshwater contributor to the Chesapeake Bay, and it is the largest river system where Salmonella diversity has been studied. Rainfall and subsequent high river discharge rates were the greatest indicators of Salmonella presence in the Susquehanna and its tributaries. Several Salmonella serovars were identified, including eight commonly associated with foodborne illness. Many clinically important serovars were present at a low frequency within individual samples and so could not be detected by conventional culture methods. The technologies employed here reveal an average of three serovars in a 10-liter sample of water and up to 10 serovars in a single sample.

Recent Advances in Our Understanding of the Diversity and Roles of Chaperone-Usher Fimbriae in Facilitating Salmonella Host and Tissue Tropism

Salmonella enterica is one of the most diverse and successful pathogens, representing a species with >2,600 serovars with a variety of adaptations that enable colonization and infection of a wide range of hosts. Fimbriae, thin hair-like projections that cover the surface of Salmonella, are thought to be the primary organelles that mediate Salmonella's interaction with, and adherence to, the host intestinal epithelium, representing an important step in the infection process. The recent expansion in genome sequencing efforts has enabled the discovery of novel fimbriae, thereby providing new perspectives on fimbrial diversity and distribution among a broad number of serovars. In this review, we provide an updated overview of the evolutionary events that shaped the Salmonella chaperone-usher fimbriome in light of recent phylogenetic studies describing the population structure of Salmonella enterica. Furthermore, we discuss the complexities of the chaperone-usher fimbriae-mediated host-pathogen interactions and the apparent redundant roles of chaperone-usher fimbriae in host and tissue tropism.

Emerging Applications of Machine Learning in Food Safety

Food safety continues to threaten public health. Machine learning holds potential in leveraging large, emerging data sets to improve the safety of the food supply and mitigate the impact of food safety incidents. Foodborne pathogen genomes and novel data streams, including text, transactional, and trade data, have seen emerging applications enabled by a machine learning approach, such as prediction of antibiotic resistance, source attribution of pathogens, and foodborne outbreak detection and risk assessment. In this article, we provide a gentle introduction to machine learning in the context of food safety and an overview of recent developments and applications. With many of these applications still in their nascence, general and domain-specific pitfalls and challenges associated with machine learning have begun to be recognized and addressed, which are critical to prospective use and future deployment of large data sets and their associated machine learning models for food safety applications.

Metagenomics-Based Approach to Source-Attribution of Antimicrobial Resistance Determinants - Identification of Reservoir Resistome Signatures

Metagenomics can unveil the genetic content of the total microbiota in different environments, such as food products and the guts of humans and livestock. It is therefore considered of great potential to investigate the transmission of foodborne hazards as part of source-attribution studies. Source-attribution of antimicrobial resistance (AMR) has traditionally relied on pathogen isolation, while metagenomics allows investigating the full span of AMR determinants. In this study, we hypothesized that the relative abundance of fecal resistome components can be associated with specific reservoirs, and that resistomes can be used for AMR source-attribution. We used shotgun-sequences from fecal samples of pigs, broilers, turkeys- and veal calves collected across Europe, and fecal samples from humans occupationally exposed to livestock in one country (pig slaughterhouse workers, pig and broiler farmers). We applied both hierarchical and flat forms of the supervised classification ensemble algorithm Random Forests to classify resistomes into corresponding reservoir classes. We identified country-specific and -independent AMR determinants, and assessed the impact of country-specific determinants when attributing AMR resistance in humans. Additionally, we performed a similarity percentage analysis with the full spectrum of AMR determinants to identify resistome signatures for the different reservoirs. We showed that the number of AMR determinants necessary to attribute a resistome into the correct reservoir increases with a larger reservoir heterogeneity, and that the impact of country-specific resistome signatures on prediction varies between countries. We predicted a higher occupational exposure to AMR determinants among workers exposed to pigs than among those exposed to broilers. Additionally, results suggested that AMR exposure on pig farms was higher than in pig slaughterhouses. Human resistomes were more similar to pig and veal calves’ resistomes than to those of broilers and turkeys, and the majority of these resistome dissimilarities can be explained by a small set of AMR determinants. We identified resistome signatures for each individual reservoir, which include AMR determinants significantly associated with on-farm antimicrobial use. We attributed human resistomes to different livestock reservoirs using Random Forests, which allowed identifying pigs as a potential source of AMR in humans. This study thus demonstrates that it is possible to apply metagenomics in AMR source-attribution.

Prevalence and phenotypic characterization of Salmonella enterica isolates from three species of wild marine turtles in Grenada, West Indies

BACKGROUND AND AIM

Salmonella enterica causes enteric disease in mammals and may potentially be transmitted from marine turtles that shed the pathogen in the environment. Marine turtle-associated human salmonellosis is a potential public health concern in Grenada, as the island supports populations of leatherback turtles (Dermochelys coriacea), hawksbill turtles (Eretmochelys imbricata), and green turtles (Chelonia mydas) that interface with veterinarians and conservation workers, the local population, and the thousands of visitors that frequent the island yearly. To date, the prevalence of S. enterica has only been examined in a small subset of marine turtles in the Caribbean and no studies have been conducted in Grenada. The aim of this study was to quantify the prevalence of S. enterica in leatherback, hawksbill and green turtles in Grenada, characterize phenotypes and DNA profiles, and explore the potential risk to human health in the region.

MATERIALS AND METHODS

A total of 102 cloacal swabs were obtained from nesting leatherback turtles and foraging hawksbill and green turtles. Samples were cultured on enrichment and selective media and isolates were phenotypically characterized using serotyping, pulsed-phase gel electrophoresis, and antibiotic susceptibility. Enrichment broths were additionally screened by polymerase chain reaction (PCR) using S. enterica-specific primers.

RESULTS

S. enterica was cultured from 15/57 (26.3%) leatherback turtles, 0/28 hawksbill, and 0/17 green turtles. This included S. enterica serovars Montevideo, S. I:4,5,12:i:-, Salmonella Typhimurium, Salmonella Newport, S. I:6,7:-:-, and S. I:4,5,12:-:-. Five/15 leatherback turtles carried multiple serovars. Eight pulsotype groups were identified with multiple clustering; however, there was no clear association between pulsotype group and serotype profile. Five/71 isolates showed resistance to streptomycin or ampicillin. Twenty-one/57 leatherback turtles, 14/28 hawksbill turtles, and 8/17 green turtles tested positive for S. enterica by quantitative PCR.

CONCLUSION

Nesting leatherback turtles actively shed S. enterica and poses a risk for zoonosis; however, the presence of viable pathogen in green and hawksbill species is unclear. These findings help elucidate the role of marine turtles as potential sources of zoonotic S. enterica and provide baseline data for one health research in Grenada and the wider Caribbean region.

Systematic evaluation of supervised machine learning for sample origin prediction using metagenomic sequencing data

Background

The advent of metagenomic sequencing provides microbial abundance patterns that can be leveraged for sample origin prediction. Supervised machine learning classification approaches have been reported to predict sample origin accurately when the origin has been previously sampled. Using metagenomic datasets provided by the 2019 CAMDA challenge, we evaluated the influence of variable technical, analytical and machine learning approaches for result interpretation and novel source prediction.

Results

Comparison between 16S rRNA amplicon and shotgun sequencing approaches as well as metagenomic analytical tools showed differences in normalized microbial abundance, especially for organisms present at low abundance. Shotgun sequence data analyzed using Kraken2 and Bracken, for taxonomic annotation, had higher detection sensitivity. As classification models are limited to labeling pre-trained origins, we took an alternative approach using Lasso-regularized multivariate regression to predict geographic coordinates for comparison. In both models, the prediction errors were much higher in Leave-1-city-out than in 10-fold cross validation, of which the former realistically forecasted the increased difficulty in accurately predicting samples from new origins. This challenge was further confirmed when applying the model to a set of samples obtained from new origins. Overall, the prediction performance of the regression and classification models, as measured by mean squared error, were comparable on mystery samples. Due to higher prediction error rates for samples from new origins, we provided an additional strategy based on prediction ambiguity to infer whether a sample is from a new origin. Lastly, we report increased prediction error when data from different sequencing protocols were included as training data.

Conclusions

Herein, we highlight the capacity of predicting sample origin accurately with pre-trained origins and the challenge of predicting new origins through both regression and classification models. Overall, this work provides a summary of the impact of sequencing technique, protocol, taxonomic analytical approaches, and machine learning approaches on the use of metagenomics for prediction of sample origin.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13062-020-00287-y.

Salmonella identified in pigs in Kenya and Malawi reveals the potential for zoonotic transmission in emerging pork markets

Salmonella is a major cause of foodborne disease globally. Pigs can carry and shed non-typhoidal Salmonella (NTS) asymptomatically, representing a significant reservoir for these pathogens. To investigate Salmonella carriage by African domestic pigs, faecal and mesenteric lymph node samples were taken at slaughter in Nairobi, Busia (Kenya) and Chikwawa (Malawi) between October 2016 and May 2017. Selective culture, antisera testing and whole genome sequencing were performed on samples from 647 pigs; the prevalence of NTS carriage was 12.7% in Busia, 9.1% in Nairobi and 24.6% in Chikwawa. Two isolates of S. Typhimurium ST313 were isolated, but were more closely related to ST313 isolates associated with gastroenteritis in the UK than bloodstream infection in Africa. The discovery of porcine NTS carriage in Kenya and Malawi reveals potential for zoonotic transmission of diarrhoeal strains to humans in these countries, but not for transmission of clades specifically associated with invasive NTS disease in Africa.

Food Safety Genomics and Connections to One Health and the Clinical Microbiology Laboratory

This article describes the potential for one health surveillance of foodborne pathogens and disease using the revolutionary methodologies of whole genome sequencing. Whole genome sequencing of viral and bacterial pathogens is a natural fit to a one health perspective because these pathogens reside and are shared by humans, animals, and the environment and their genomes are compared easily regardless of where or from what host the pathogen was isolated. A genome provides a huge amount of data that can be analyzed for numerous applications. Sharing data coordinates surveillance efforts across the various disciplines.

Application of Whole-Genome Sequences and Machine Learning in Source Attribution of Salmonella Typhimurium

Prevention of the emergence and spread of foodborne diseases is an important prerequisite for the improvement of public health. Source attribution models link sporadic human cases of a specific illness to food sources and animal reservoirs. With the next generation sequencing technology, it is possible to develop novel source attribution models. We investigated the potential of machine learning to predict the animal reservoir from which a bacterial strain isolated from a human salmonellosis case originated based on whole-genome sequencing. Machine learning methods recognize patterns in large and complex data sets and use this knowledge to build models. The model learns patterns associated with genetic variations in bacteria isolated from the different animal reservoirs. We selected different machine learning algorithms to predict sources of human salmonellosis cases and trained the model with Danish Salmonella Typhimurium isolates sampled from broilers (n = 34), cattle (n = 2), ducks (n = 11), layers (n = 4), and pigs (n = 159). Using cgMLST as input features, the model yielded an average accuracy of 0.783 (95% CI: 0.77-0.80) in the source prediction for the random forest and 0.933 (95% CI: 0.92-0.94) for the logit boost algorithm. Logit boost algorithm was most accurate (valid accuracy: 92%, CI: 0.8706-0.9579) and predicted the origin of 81% of the domestic sporadic human salmonellosis cases. The most important source was Danish produced pigs (53%) followed by imported pigs (16%), imported broilers (6%), imported ducks (2%), Danish produced layers (2%), Danish produced cattle and imported cattle (<1%) while 18% was not predicted. Machine learning has potential for improving source attribution modeling based on sequence data. Results of such models can inform risk managers to identify and prioritize food safety interventions.

AB_SA: Accessory genes-Based Source Attribution - tracing the source of Salmonella enterica Typhimurium environmental strains

The partitioning of pathogenic strains isolated in environmental or human cases to their sources is challenging. The pathogens usually colonize multiple animal hosts, including livestock, which contaminate the food-production chain and the environment (e.g. soil and water), posing an additional public-health burden and major challenges in the identification of the source. Genomic data opens up new opportunities for the development of statistical models aiming to indicate the likely source of pathogen contamination. Here, we propose a computationally fast and efficient multinomial logistic regression source-attribution classifier to predict the animal source of bacterial isolates based on 'source-enriched' loci extracted from the accessory-genome profiles of a pangenomic dataset. Depending on the accuracy of the model's self-attribution step, the modeller selects the number of candidate accessory genes that best fit the model for calculating the likelihood of (source) category membership. The Accessory genes-Based Source Attribution (AB_SA) method was applied to a dataset of strains of Salmonella enterica Typhimurium and its monophasic variant (S. enterica 1,4,[5],12:i:-). The model was trained on 69 strains with known animal-source categories (i.e. poultry, ruminant and pig). The AB_SA method helped to identify 8 genes as predictors among the 2802 accessory genes. The self-attribution accuracy was 80 %. The AB_SA model was then able to classify 25 of the 29 S. enterica Typhimurium and S. enterica 1,4,[5],12:i:- isolates collected from the environment (considered to be of unknown source) into a specific category (i.e. animal source), with more than 85 % of probability. The AB_SA method herein described provides a user-friendly and valuable tool for performing source-attribution studies in only a few steps. AB_SA is written in R and freely available at https://github.com/lguillier/AB_SA.

Evolution of Salmonella enterica serotype Typhimurium driven by anthropogenic selection and niche adaptation

Salmonella enterica serotype Typhimurium (S. Typhimurium) is a leading cause of gastroenteritis and bacteraemia worldwide, and a model organism for the study of host-pathogen interactions. Two S. Typhimurium strains (SL1344 and ATCC14028) are widely used to study host-pathogen interactions, yet genotypic variation results in strains with diverse host range, pathogenicity and risk to food safety. The population structure of diverse strains of S. Typhimurium revealed a major phylogroup of predominantly sequence type 19 (ST19) and a minor phylogroup of ST36. The major phylogroup had a population structure with two high order clades (α and β) and multiple subclades on extended internal branches, that exhibited distinct signatures of host adaptation and anthropogenic selection. Clade α contained a number of subclades composed of strains from well characterized epidemics in domesticated animals, while clade β contained multiple subclades associated with wild avian species. The contrasting epidemiology of strains in clade α and β was reflected by the distinct distribution of antimicrobial resistance (AMR) genes, accumulation of hypothetically disrupted coding sequences (HDCS), and signatures of functional diversification. These observations were consistent with elevated anthropogenic selection of clade α lineages from adaptation to circulation in populations of domesticated livestock, and the predisposition of clade β lineages to undergo adaptation to an invasive lifestyle by a process of convergent evolution with of host adapted Salmonella serotypes. Gene flux was predominantly driven by acquisition and recombination of prophage and associated cargo genes, with only occasional loss of these elements. The acquisition of large chromosomally-encoded genetic islands was limited, but notably, a feature of two recent pandemic clones (DT104 and monophasic S. Typhimurium ST34) of clade α (SGI-1 and SGI-4).

Rapid in-country sequencing of whole virus genomes to inform rabies elimination programmes

Genomic surveillance is an important aspect of contemporary disease management but has yet to be used routinely to monitor endemic disease transmission and control in low- and middle-income countries. Rabies is an almost invariably fatal viral disease that causes a large public health and economic burden in Asia and Africa, despite being entirely vaccine preventable. With policy efforts now directed towards achieving a global goal of zero dog-mediated human rabies deaths by 2030, establishing effective surveillance tools is critical. Genomic data can provide important and unique insights into rabies spread and persistence that can direct control efforts. However, capacity for genomic research in low- and middle-income countries is held back by limited laboratory infrastructure, cost, supply chains and other logistical challenges. Here we present and validate an end-to-end workflow to facilitate affordable whole genome sequencing for rabies surveillance utilising nanopore technology. We used this workflow in Kenya, Tanzania and the Philippines to generate rabies virus genomes in two to three days, reducing costs to approximately £60 per genome. This is over half the cost of metagenomic sequencing previously conducted for Tanzanian samples, which involved exporting samples to the UK and a three- to six-month lag time. Ongoing optimization of workflows are likely to reduce these costs further. We also present tools to support routine whole genome sequencing and interpretation for genomic surveillance. Moreover, combined with training workshops to empower scientists in-country, we show that local sequencing capacity can be readily established and sustainable, negating the common misperception that cutting-edge genomic research can only be conducted in high resource laboratories. More generally, we argue that the capacity to harness genomic data is a game-changer for endemic disease surveillance and should precipitate a new wave of researchers from low- and middle-income countries.

A guide to machine learning for bacterial host attribution using genome sequence data

With the ever-expanding number of available sequences from bacterial genomes, and the expectation that this data type will be the primary one generated from both diagnostic and research laboratories for the foreseeable future, then there is both an opportunity and a need to evaluate how effectively computational approaches can be used within bacterial genomics to predict and understand complex phenotypes, such as pathogenic potential and host source. This article applied various quantitative methods such as diversity indexes, pangenome-wide association studies (GWAS) and dimensionality reduction techniques to better understand the data and then compared how well unsupervised and supervised machine learning (ML) methods could predict the source host of the isolates. The study uses the example of the pangenomes of 1203 Salmonella enterica serovar Typhimurium isolates in order to predict 'host of isolation' using these different methods. The article is aimed as a review of recent applications of ML in infection biology, but also, by working through this specific dataset, it allows discussion of the advantages and drawbacks of the different techniques. As with all such sub-population studies, the biological relevance will be dependent on the quality and diversity of the input data. Given this major caveat, we show that supervised ML has the potential to add real value to interpretation of bacterial genomic data, as it can provide probabilistic outcomes for important phenotypes, something that is very difficult to achieve with the other methods.

Four European Salmonella Typhimurium datasets collected to develop WGS-based source attribution methods

Zoonotic Salmonella causes millions of human salmonellosis infections worldwide each year. Information about the source of the bacteria guides risk managers on control and preventive strategies. Source attribution is the effort to quantify the number of sporadic human cases of a specific illness to specific sources and animal reservoirs. Source attribution methods for Salmonella have so far been based on traditional wet-lab typing methods. With the change to whole genome sequencing there is a need to develop new methods for source attribution based on sequencing data. Four European datasets collected in Denmark (DK), Germany (DE), the United Kingdom (UK) and France (FR) are presented in this descriptor. The datasets contain sequenced samples of Salmonella Typhimurium and its monophasic variants isolated from human, food, animal and the environment. The objective of the datasets was either to attribute the human salmonellosis cases to animal reservoirs or to investigate contamination of the environment by attributing the environmental isolates to different animal reservoirs.