Concordance of SNP- and allele-based typing workflows in the context of a large-scale international Salmonella Enteritidis outbreak investigation

Lessons from 5 Years of Routine Whole-Genome Sequencing for Epidemiologic Surveillance of Shiga Toxin-Producing Escherichia coli, France, 2018-2022

Whole-genome sequencing (WGS) is routine for surveillance of Shiga toxin-producing Escherichia coli human isolates in France. Protocols use EnteroBase hierarchical clustering at <5 allelic differences (HC5) as screening for cluster detection. We assessed current implementation after 5 years for 1,002 sequenced isolates. From genomic distances of serotypes O26:H11, O157:H7, O80:H2, and O103:H2, we determined statistical thresholds for cluster determination and compared those with HC5 clusters. Thresholds varied by serotype, 5-16 allelic distances and 15-20 single-nucleotide polymorphisms, showing limits of a single-threshold approach. We confirmed validity of HC5 screening for 3 serotypes because statistical thresholds had limited effect on isolate clustering (high sensitivity and specificity). For O80:H2, results suggest that HC5 is less reliable, and other approaches should be explored. Public health officials should regularly assess WGS used for Shiga toxin-producing E. coli surveillance to account for serotype and genomic evolution and to interpret WGS-linked isolates in light of epidemiologic data.

Molecular epidemiology of Salmonella Enteritidis in humans and animals in Spain

Salmonella Enteritidis, the most prevalent serovar-causing human gastroenteritis, has been traditionally linked to poultry sources. Although antimicrobial resistance (AMR) is not common in this serovar, increasing levels of resistance to fluoroquinolones and ampicillin have been reported in the last few years. Here, 298 isolates retrieved from different sources (human, livestock, wildlife, food, and environment) and years (2002-2021) in Spain were analyzed to evaluate their diversity, the distribution of AMR-conferring genes (ARGs), and mutations and reconstruct the epidemiology of infection due to this serovar. Isolates were clustered in two major clades (I and II), with strains in clade I (including 61.5% of all human isolates) displaying a pan-susceptible phenotype and not carrying AMR determinants. In contrast, clade II included 80.7% of isolates from animal/food/environmental sources, with the majority (69.8%) harboring mutations in the quinolone resistance determinant regions (QRDR). ARGs, although rare, were mostly found in clade II strains that also carried plasmid replicons, among which IncX1 was the most common. Although higher levels of phenotypic resistance were found in animal isolates, extended-spectrum beta-lactamase, plasmid-mediated AmpC, and carbapenemase-encoding genes were only found among human isolates. In summary, the majority of human and animal isolates from Spanish sources in our collection were classified in different phylogenetic branches, suggesting that additional sources are contributing to the occurrence of foodborne infections in Spain. Furthermore, the different distributions of virulence factors and ARGs in isolates from different sources and their association with specific plasmids suggest the presence of different dynamics contributing to the selection of resistant strains.

Genomic epidemiology of Salmonella Enteritidis human infections in the Netherlands, 2019 to 2023

Salmonella enterica serotype Enteritidis (SE) is a common foodborne pathogen that can cause human salmonellosis. Identifying closely related cases is essential to control the pathogen through, e.g. outbreak investigation, but it is often challenging due to the low genetic diversity of SE, particularly with traditional typing methods. This study aimed to investigate the population structure of SE genomes collected during routine surveillance in the Netherlands using whole-genome sequencing (WGS), their clustering, temporal distribution and the association between epidemiological and phenotypic antimicrobial resistance (AMR) factors and the persistence of SE clusters. We also investigated the distribution of genotypic AMR markers among these isolates. The study collection comprised 1,669 unique SE isolates from human infections collected from Dutch surveillance between 2019 and 2023, and their relatedness was derived using core-genome multi-locus sequence typing and Hamming distances. Based on the results, the 216 clusters comprised 1,085 sequences, in addition to 584 sequences depicted as singletons. These clusters predominantly fell within three major lineages, of which two were the previously described Global and Atlantic lineages. Of these clusters, approximately a third persisted for more than 1 year during the 5-year study period. However, no statistically significant associations were found between epidemiological factors, such as age, gender and travel history, or phenotypic AMR and the persistence of SE clusters. The most common AMR genetic markers observed were related to antimicrobial classes of (fluor)quinolones, β-lactamases and aminoglycosides. This study provides a better understanding of the genomic epidemiology of SE in the Netherlands based on WGS. Further analysis that includes samples from the food-chain supply, along with higher resolution methods during a post-Coronavirus Disease of 2019 (COVID-19) period, may provide more insights into the possible causes of the persistence of SE clusters.

Salmonella associated with agricultural animals exhibit diverse evolutionary rates and show evidence of recent clonal expansion

Most foodborne salmonellosis outbreaks are linked to agricultural animal products with a few serovars accounting for most Salmonella isolated from specific animal products, suggesting an adaptation to the corresponding animal hosts and their respective environments. Here, we utilized whole-genome sequence (WGS) data to analyze the evolution and population genetics of seven serovars frequently isolated from ground beef (Montevideo, Cerro, and Dublin), chicken (Kentucky, Infantis, and Enteritidis), and turkey (Reading) in the United States. In addition, publicly available metadata were used to characterize major clades within each serovar with regard to public health significance. Except for Dublin, all serovars were polyphyletic, comprising 2-6 phylogenetic groups. Further partitioning of the phylogenies identified 25 major clades, including 12 associated with animal or environmental niches. These 12 clades differed in evolutionary parameters (e.g., substitution rates) as well as public health relevant characteristics (e.g., association with human illness, antimicrobial resistance). Overall, our results highlight several critical trends: (i) the Salmonella generation time appears to be more dependent on source than serovar and (ii) all serovars contain clades and sub-clades that are estimated to have emerged after the year 1940 and that are enriched for isolates associated with humans, agricultural animals, antimicrobial resistance (AMR), and/or specific geographical regions. These findings suggest that serotyping alone does not provide enough resolution to differentiate isolates that may have evolved independently, present distinct geographic distribution and host association, and possibly have distinct public health significance.

IMPORTANCE

Non-typhoidal Salmonella are major foodborne bacterial pathogens estimated to cause more than one million illnesses, thousands of hospitalizations, and hundreds of deaths annually in the United States. More than 70% of Salmonella outbreaks in the United States have been associated with agricultural animals. Certain serovars include persistent strains that have repeatedly contaminated beef, chicken, and turkey, causing outbreaks and sporadic cases over many years. These persistent strains represent a particular challenge to public health, as they are genetically clonal and widespread, making it difficult to differentiate distinct outbreak and contamination events using whole-genome sequence (WGS)-based subtyping methods (e.g., core genome allelic typing). Our results indicate that a phylogenetic approach is needed to investigate persistent strains and suggest that the association between a Salmonella serovar and an agricultural animal is driven by the expansion of clonal subtypes that likely became adapted to specific animals and associated environments.

Use of whole genome sequencing for surveillance and control of foodborne diseases: status quo and quo vadis

Improvements in sequencing quality, availability, speed and costs results in an increased presence of genomics in infectious disease applications. Nevertheless, there are still hurdles in regard to the optimal use of WGS for public health purposes. Here, we discuss the current state ("status quo") and future directions ("quo vadis") based on literature regarding the use of genomics in surveillance, hazard characterization and source attribution of foodborne pathogens. The future directions include the application of new techniques, such as machine learning and network approaches that may overcome the current shortcomings. These include the use of fixed genomic distances in cluster delineation, disentangling similarity or lack thereof in source attribution, and difficulties ascertaining function in hazard characterization. Although, the aforementioned methods can relatively easily be applied technically, an overarching challenge is the inference and biological/epidemiological interpretation of these large amounts of high-resolution data. Understanding the context in terms of bacterial isolate and host diversity allows to assess the level of representativeness in regard to sources and isolates in the dataset, which in turn defines the level of certainty associated with defining clusters, sources and risks. This also marks the importance of metadata (clinical, epidemiological, and biological) when using genomics for public health purposes.

Evaluation of whole and core genome multilocus sequence typing allele schemes for Salmonella enterica outbreak detection in a national surveillance network, PulseNet USA

Salmonella enterica is a leading cause of bacterial foodborne and zoonotic illnesses in the United States. For this study, we applied four different whole genome sequencing (WGS)-based subtyping methods: high quality single-nucleotide polymorphism (hqSNP) analysis, whole genome multilocus sequence typing using either all loci [wgMLST (all loci)] and only chromosome-associated loci [wgMLST (chrom)], and core genome multilocus sequence typing (cgMLST) to a dataset of isolate sequences from 9 well-characterized Salmonella outbreaks. For each outbreak, we evaluated the genomic and epidemiologic concordance between hqSNP and allele-based methods. We first compared pairwise genomic differences using all four methods. We observed discrepancies in allele difference ranges when using wgMLST (all loci), likely caused by inflated genetic variation due to loci found on plasmids and/or other mobile genetic elements in the accessory genome. Therefore, we excluded wgMLST (all loci) results from any further comparisons in the study. Then, we created linear regression models and phylogenetic tanglegrams using the remaining three methods. K-means analysis using the silhouette method was applied to compare the ability of the three methods to partition outbreak and sporadic isolate sequences. Our results showed that pairwise hqSNP differences had high concordance with cgMLST and wgMLST (chrom) allele differences. The slopes of the regressions for hqSNP vs. allele pairwise differences were 0.58 (cgMLST) and 0.74 [wgMLST (chrom)], and the slope of the regression was 0.77 for cgMLST vs. wgMLST (chrom) pairwise differences. Tanglegrams showed high clustering concordance between methods using two statistical measures, the Baker's gamma index (BGI) and cophenetic correlation coefficient (CCC), where 9/9 (100%) of outbreaks yielded BGI values ≥ 0.60 and CCCs were ≥ 0.97 across all nine outbreaks and all three methods. K-means analysis showed separation of outbreak and sporadic isolate groups with average silhouette widths ≥ 0.87 for outbreak groups and ≥ 0.16 for sporadic groups. This study demonstrates that Salmonella isolates clustered in concordance with epidemiologic data using three WGS-based subtyping methods and supports using cgMLST as the primary method for national surveillance of Salmonella outbreak clusters.

Highly drug resistant clone of Salmonella Kentucky ST198 in clinical infections and poultry in Zimbabwe

A highly multidrug-resistant strain of Salmonella enterica serotype Kentucky (S. Kentucky) of sequence type (ST)198 emerged in North Africa and has since spread widely. To investigate the genetic diversity and phylogenetic relationship of S. Kentucky in Zimbabwe and identify potential sources of infection, the whole-genome sequence of 37 S. Kentucky strains isolated from human clinical infections and from poultry farms between 2017 and 2020 was determined. Of 37 S. Kentucky isolates, 36 were ST198 and one was ST152. All ST198 isolates had between six and fifteen antimicrobial resistance (AMR) genes, and 92% carried at least ten AMRs. All ST198 isolates harbored the Salmonella genomic island K-Israel variant (SGI1-KIV) integrated into the chromosome with aac(3)-ld, aac(6)-laa, aadA7, blaTEM-1, sul1, and tetA genes, with occasional sporadic loss of one or more genes noted from five isolates. All ST198 isolates also had mutations in the quinolone resistance-determining region of the gyrA and parC genes. The blaCTX-M-14.1 and fosA3 genes were present in 92% of the ST198 isolates, conferring resistance to extended-spectrum cephalosporins and fosfomycin, respectively, were present on an IncHI2 plasmid with the aadA2b, aadA1, aph(3')-Ib, aph(6')-Id, cmlA1 and sul3 AMR genes. S. Kentucky ST198 isolates from Zimbabwe formed a closely related phylogenetic clade that emerged from a previously reported global epidemic population. The close genetic relationship and population structure of the human clinical and poultry isolates of ST198 in Zimbabwe are consistent with poultry being an important source of highly resistant strains of S. Kentucky in Zimbabwe.

Enteric fever cluster identification in South Africa using genomic surveillance of Salmonella enterica serovar Typhi

The National Institute for Communicable Diseases in South Africa participates in national laboratory-based surveillance for human isolates of Salmonella species. Laboratory analysis includes whole-genome sequencing (WGS) of isolates. We report on WGS-based surveillance of Salmonella enterica serovar Typhi (Salmonella Typhi) in South Africa from 2020 through 2021. We describe how WGS analysis identified clusters of enteric fever in the Western Cape Province of South Africa and describe the epidemiological investigations associated with these clusters. A total of 206 Salmonella Typhi isolates were received for analysis. Genomic DNA was isolated from bacteria and WGS was performed using Illumina NextSeq technology. WGS data were investigated using multiple bioinformatics tools, including those available at the Centre for Genomic Epidemiology, EnteroBase and Pathogenwatch. Core-genome multilocus sequence typing was used to investigate the phylogeny of isolates and identify clusters. Three major clusters of enteric fever were identified in the Western Cape Province; cluster one (n=11 isolates), cluster two (n=13 isolates), and cluster three (n=14 isolates). To date, no likely source has been identified for any of the clusters. All isolates associated with the clusters, showed the same genotype (4.3.1.1.EA1) and resistome (antimicrobial resistance genes: bla _TEM-1B, catA1, sul1, sul2, dfrA7). The implementation of genomic surveillance of Salmonella Typhi in South Africa has enabled rapid detection of clusters indicative of possible outbreaks. Cluster identification allows for targeted epidemiological investigations and a timely, coordinated public health response.

A Snapshot of the Genetic Diversity of Salmonella Enteritidis Population Involved in Human Infections in Romania Taken in the European Epidemiological Context

In the absence of consistent national molecular typing data to enhance the surveillance of Salmonella Enteritidis, it was considered useful to collect baseline information on the genetic diversity and antibiotic susceptibility of strains isolated in Romania between January 2016 and April 2020 and compare them to strains described in major international outbreaks of the same period. A collection of 245 clinical isolates were genotyped by a standardised multiple-locus variable-number of tandem repeats analysis (MLVA) 5-loci protocol and screened for antimicrobial resistance against 15 compounds. Twenty strains were further subjected to whole genome sequencing (WGS) and compared to epidemiologically relevant high-throughput sequencing data available in European databases. Twenty-seven MLVA genotypes were identified, of which three, commonly reported in Europe between 2016–2020, covered 72% of the collection. Antibiotic resistance was detected in 30% of the strains, with resistance to nalidixic acid and ciprofloxacin as the most common phenotype, and also associated with two prevalent MLVA clones. WGS-derived multilocus sequence typing (MLST) revealed a single sequence type (ST11) further resolved into 10 core-genome MLST complex types. The minimum spanning tree constructed from the cgMLST data clustered Romanian and international strains, which shared more than 95% of the core genes, revealing links with a contemporaneous multi-country outbreak. This study could be regarded as a forerunner to the advent of using this integrative approach in the public health practice at a national level and thus contribute to the concerted actions at a European level to stop outbreaks.

SOSPCNN: Structurally Optimized Stochastic Pooling Convolutional Neural Network for Tetralogy of Fallot recognition

Aim

This study proposes a new artificial intelligence model based on cardiovascular computed tomography for more efficient and precise recognition of Tetralogy of Fallot (TOF).

Methods

Our model is a structurally optimized stochastic pooling convolutional neural network (SOSPCNN), which combines stochastic pooling, structural optimization, and convolutional neural network. In addition, multiple-way data augmentation is used to overcome overfitting. Grad-CAM is employed to provide explainability to the proposed SOSPCNN model. Meanwhile, both desktop and web apps are developed based on this SOSPCNN model.

Results

The results on ten runs of 10-fold cross-validation show that our SOSPCNN model yields a sensitivity of 92.25±2.19, a specificity of 92.75±2.49, a precision of 92.79±2.29, an accuracy of 92.50±1.18, an F1 score of 92.48±1.17, an MCC of 85.06±2.38, an FMI of 92.50±1.17, and an AUC of 0.9587.

Conclusion

The SOSPCNN method performed better than three state-of-the-art TOF recognition approaches.

Decentralized Investigation of Bacterial Outbreaks Based on Hashed cgMLST

Whole-genome sequencing (WGS)-based outbreak investigation has proven to be a valuable method for the surveillance of bacterial pathogens. Its utility has been successfully demonstrated using both gene-by-gene (cgMLST or wgMLST) and single-nucleotide polymorphism (SNP)-based approaches. Among the obstacles of implementing a WGS-based routine surveillance is the need for an exchange of large volumes of sequencing data, as well as a widespread reluctance to share sequence and metadata in public repositories, together with a lacking standardization of suitable bioinformatic tools and workflows. To address these issues, we present chewieSnake, an intuitive and simple-to-use cgMLST workflow. ChewieSnake builds on the allele calling software chewBBACA and extends it by the concept of allele hashing. The resulting hashed allele profiles can be readily compared between laboratories without the need of a central allele nomenclature. The workflow fully automates the computation of the allele distance matrix, cluster membership, and phylogeny and summarizes all important findings in an interactive HTML report. Furthermore, chewieSnake can join allele profiles generated at different laboratories and identify shared clusters, including a stable and intercommunicable cluster nomenclature, thus facilitating a joint outbreak investigation. We demonstrate the feasibility of the proposed approach with a thorough method comparison using publically available sequencing data for Salmonella enterica. However, chewieSnake is readily applicable to all bacterial taxa, provided that a suitable cgMLST scheme is available. The workflow is freely available as an open-source tool and can be easily installed via conda or docker.

High-Resolution Genomic Comparisons within Salmonella enterica Serotypes Derived from Beef Feedlot Cattle: Parsing the Roles of Cattle Source, Pen, Animal, Sample Type, and Production Period

Salmonella enterica is a major foodborne pathogen, and contaminated beef products have been identified as one of the primary sources of Salmonella-related outbreaks. Pathogenicity and antibiotic resistance of Salmonella are highly serotype and subpopulation specific, which makes it essential to understand high-resolution Salmonella population dynamics in cattle. Time of year, source of cattle, pen, and sample type (i.e., feces, hide, or lymph nodes) have previously been identified as important factors influencing the serotype distribution of Salmonella (e.g., Anatum, Lubbock, Cerro, Montevideo, Kentucky, Newport, and Norwich) that were isolated from a longitudinal sampling design in a research feedlot. In this study, we performed high-resolution genomic comparisons of Salmonella isolates within each serotype using both single-nucleotide polymorphism-based maximum-likelihood phylogeny and hierarchical clustering of core-genome multilocus sequence typing. The importance of the aforementioned features in clonal Salmonella expansion was further explored using a supervised machine learning algorithm. In addition, we identified and compared the resistance genes, plasmids, and pathogenicity island profiles of the isolates within each subpopulation. Our findings indicate that clonal expansion of Salmonella strains in cattle was mainly influenced by the randomization of block and pen, as well as the origin/source of the cattle, i.e., regardless of sampling time and sample type (i.e., feces, lymph node, or hide). Further research is needed concerning the role of the feedlot pen environment prior to cattle placement to better understand carryover contributions of existing strains of Salmonella and their bacteriophages. IMPORTANCE Salmonella serotypes isolated from outbreaks in humans can also be found in beef cattle and feedlots. Virulence factors and antibiotic resistance are among the primary defense mechanisms of Salmonella, and are often associated with clonal expansion. This makes understanding the subpopulation dynamics of Salmonella in cattle critical for effective mitigation. There remains a gap in the literature concerning subpopulation dynamics within Salmonella serotypes in feedlot cattle from the beginning of feeding up until slaughter. Here, we explore Salmonella population dynamics within each serotype using core-genome phylogeny and hierarchical classifications. We used machine learning to quantitatively parse the relative importance of both hierarchical and longitudinal clustering among cattle host samples. Our results reveal that Salmonella populations in cattle are highly clonal over a 6-month study period and that clonal dissemination of Salmonella in cattle is mainly influenced spatially by experimental block and pen, as well by the geographical origin of the cattle.

HierCC: A multi-level clustering scheme for population assignments based on core genome MLST

Motivation

Routine infectious disease surveillance is increasingly based on large-scale whole-genome sequencing databases. Real-time surveillance would benefit from immediate assignments of each genome assembly to hierarchical population structures. Here we present pHierCC, a pipeline that defines a scalable clustering scheme, HierCC, based on core genome multi-locus typing that allows incremental, static, multi-level cluster assignments of genomes. We also present HCCeval, which identifies optimal thresholds for assigning genomes to cohesive HierCC clusters. HierCC was implemented in EnteroBase in 2018 and has since genotyped >530 000 genomes from Salmonella, Escherichia/Shigella, Streptococcus, Clostridioides, Vibrio and Yersinia.

Availability and implementation

https://enterobase.warwick.ac.uk/ and Source code and instructions: https://github.com/zheminzhou/pHierCC

Supplementary information

Supplementary data are available at Bioinformatics online.

Whole Genome Sequencing Applied to Pathogen Source Tracking in Food Industry: Key Considerations for Robust Bioinformatics Data Analysis and Reliable Results Interpretation

Whole genome sequencing (WGS) has arisen as a powerful tool to perform pathogen source tracking in the food industry thanks to several developments in recent years. However, the cost associated to this technology and the degree of expertise required to accurately process and understand the data has limited its adoption at a wider scale. Additionally, the time needed to obtain actionable information is often seen as an impairment for the application and use of the information generated via WGS. Ongoing work towards standardization of wet lab including sequencing protocols, following guidelines from the regulatory authorities and international standardization efforts make the technology more and more accessible. However, data analysis and results interpretation guidelines are still subject to initiatives coming from distinct groups and institutions. There are multiple bioinformatics software and pipelines developed to handle such information. Nevertheless, little consensus exists on a standard way to process the data and interpret the results. Here, we want to present the constraints we face in an industrial setting and the steps we consider necessary to obtain high quality data, reproducible results and a robust interpretation of the obtained information. All of this, in a time frame allowing for data-driven actions supporting factories and their needs.

Whole-genome epidemiology links phage-mediated acquisition of a virulence gene to the clonal expansion of a pandemic Salmonella enterica serovar Typhimurium clone

Epidemic and pandemic clones of bacterial pathogens with distinct characteristics continually emerge, replacing those previously dominant through mechanisms that remain poorly characterized. Here, whole-genome-sequencing-powered epidemiology linked horizontal transfer of a virulence gene, sopE, to the emergence and clonal expansion of a new epidemic Salmonella enterica serovar Typhimurium (S. Typhimurium) clone. The sopE gene is sporadically distributed within the genus Salmonella and rare in S. enterica Typhimurium lineages, but was acquired multiple times during clonal expansion of the currently dominant pandemic monophasic S. Typhimurium sequence type (ST) 34 clone. Ancestral state reconstruction and time-scaled phylogenetic analysis indicated that sopE was not present in the common ancestor of the epidemic clade, but later acquisition resulted in increased clonal expansion of sopE-containing clones that was temporally associated with emergence of the epidemic, consistent with increased fitness. The sopE gene was mainly associated with a temperate bacteriophage mTmV, but recombination with other bacteriophage and apparent horizontal gene transfer of the sopE gene cassette resulted in distribution among at least four mobile genetic elements within the monophasic S. enterica Typhimurium ST34 epidemic clade. The mTmV prophage lysogenic transfer to other S. enterica serovars in vitro was limited, but included the common pig-associated S. enterica Derby (S. Derby). This may explain mTmV in S. Derby co-circulating on farms with monophasic S. Typhimurium ST34, highlighting the potential for further transfer of the sopE virulence gene in nature. We conclude that whole-genome epidemiology pinpoints potential drivers of evolutionary and epidemiological dynamics during pathogen emergence, and identifies targets for subsequent research in epidemiology and bacterial pathogenesis.

Evaluation of an ensemble-based distance statistic for clustering MLST datasets using epidemiologically defined clusters of cyclosporiasis

Outbreaks of cyclosporiasis, a food-borne illness caused by the coccidian parasite Cyclospora cayetanensis have increased in the USA in recent years, with approximately 2300 laboratory-confirmed cases reported in 2018. Genotyping tools are needed to inform epidemiological investigations, yet genotyping Cyclospora has proven challenging due to its sexual reproductive cycle which produces complex infections characterized by high genetic heterogeneity. We used targeted amplicon deep sequencing and a recently described ensemble-based distance statistic that accommodates heterogeneous (mixed) genotypes and specimens with partial genotyping data, to genotype and cluster 648 C. cayetanensis samples submitted to CDC in 2018. The performance of the ensemble was assessed by comparing ensemble-identified genetic clusters to analogous clusters identified independently based on common food exposures. Using these epidemiologic clusters as a gold standard, the ensemble facilitated genetic clustering with 93.8% sensitivity and 99.7% specificity. Hence, we anticipate that this procedure will greatly complement epidemiologic investigations of cyclosporiasis.