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.

Levels and genotypes of Salmonella and levels of Escherichia coli in frozen ready-to-cook chicken and turkey products in England tested in 2020 in relation to an outbreak of S. Enteritidis.

Frozen reformulated (FR) breaded chicken products have previously been implicated in causing human salmonellosis. A multi-country Salmonella enterica serovar Enteritidis outbreak involving several strains with >400 reported human cases in the UK occurred in 2020. Initially S. Infantis was detected in one sample from a case home but S. Enteritidis was then also isolated using a S. Enteritidis specific PCR in combination with isolation via a Craigie-tube. This prompted a survey to examine the presence and levels of Salmonella and E. coli in ready-to-cook FR poultry products in England in 2020. From a total of 483 samples, including two from cases' homes, Salmonella was detected in 42 chicken samples, these originated from six out of 53 production plants recorded. Salmonella detection was associated with elevated levels of generic E. coli (OR = 6.63). S. Enteritidis was detected in 17 samples, S. Infantis in 25, S. Newport in four and S. Java, S. Livingstone and S. Senftenberg in one each. The highest levels of Salmonella were 54 MPN/g for S. Infantis and 28 MPN/g for S. Enteritidis; 60% of the Salmonella-positive samples had <1.0 MPN/g. S. Enteritidis was detected together with S. Infantis in five samples and with S. Livingstone in one. Where S. Enteritidis was detected with other Salmonella, the former was present at between 2 and 100-fold lower concentrations. The Salmonella contamination was homogeneously distributed amongst chicken pieces from a single pack and present in both the outer coating and inner content. The S. Enteritidis were all outbreak strains and detected in six products that were linked to four production plants which implicated a Polish origin of contamination. Despite S. Infantis being most prevalent in these products, S. Infantis from only two contemporaneous human cases in the UK fell into the same cluster as isolates detected in one product. Except for one human case falling into the same cluster as one of the S. Newport strains from the chicken, no further isolates from human cases fell into clusters with any of the other serovars detected in the chicken samples. This study found that higher E. coli levels indicated a higher probability of Salmonella contamination in FR chicken products. The results also highlight the importance of recognising co-contamination of foods with multiple Salmonella types and has provided essential information for detecting and understanding outbreaks where multiple strains are involved.

Acquisition and loss of CTX-M plasmids in Shigella species associated with MSM transmission in the UK.

Shigellosis in men who have sex with men (MSM) is caused by multidrug resistant Shigellae, exhibiting resistance to antimicrobials including azithromycin, ciprofloxacin and more recently the third-generation cephalosporins. We sequenced four blaCTX-M-27-positive MSM Shigella isolates (2018-20) using Oxford Nanopore Technologies; three S. sonnei (identified as two MSM clade 2, one MSM clade 5) and one S. flexneri 3a, to explore AMR context. All S. sonnei isolates harboured Tn7/Int2 chromosomal integrons, whereas S. flexneri 3a contained the Shigella Resistance Locus. All strains harboured IncFII pKSR100-like plasmids (67-83kbp); where present blaCTX-M-27 was located on these plasmids flanked by IS26 and IS903B, however blaCTX-M-27 was lost in S. flexneri 3a during storage between Illumina and Nanopore sequencing. IncFII AMR regions were mosaic and likely reorganised by IS26; three of the four plasmids contained azithromycin-resistance genes erm(B) and mph(A) and one harboured the pKSR100 integron. Additionally, all S. sonnei isolates possessed a large IncB/O/K/Z plasmid, two of which carried aph(3')-Ib/aph(6)-Id/sul2 and tet(A). Monitoring the transmission of mobile genetic elements with co-located AMR determinants is necessary to inform empirical treatment guidance and clinical management of MSM-associated shigellosis.

National outbreak of Shiga toxin-producing Escherichia coli O157:H7 linked to mixed salad leaves, United Kingdom, 2016.

We investigated a large outbreak of Escherichia coli O157 in the United Kingdom (UK) with 165 cases between 31 May and 29 July 2016. No linked cases were reported in other countries. Cases were predominately female (n = 128) and adult (n = 150), 66 attended hospital and nine had features of haemorrhagic uraemic syndrome. A series of epidemiological studies (case-control, case-case, ingredients-based and venue-based studies) and supply chain investigations implicated mixed salad leaves from Supplier A as the likely outbreak vehicle. Whole genome sequencing (WGS) indicated a link with strains from the Mediterranean and informed the outbreak control team to request that Supplier A cease distributing salad leaves imported from Italy. Microbiological tests of samples of salad leaves from Supplier A were negative. We were unable to confirm the source of contamination or the contaminated constituent leaf although our evidence pointed to red batavia received from Italy as the most likely vehicle. Variations in Shiga toxin-producing E.coli surveillance and diagnosis may have prevented detection of cases outside the UK and highlights a need for greater standardisation. WGS was useful in targeting investigations, but greater coverage across Europe is needed to maximise its potential.

Investigation using whole genome sequencing of a prolonged restaurant outbreak of Salmonella Typhimurium linked to the building drainage system, England, February 2015 to March 2016.

Following notification of a Salmonella enterica serovar Typhimurium gastroenteritis outbreak, we identified 82 cases linked to a restaurant with symptom onset from 12 February 2015 to 8 March 2016. Seventy-two cases had an isolate matching the nationally unique whole genome sequencing profile (single nucleotide polymorphism (SNP) address: 1.1.1.124.395.395). Interviews established exposure to the restaurant and subsequent case-control analysis identified an association with eating carvery buffet food (adjusted odds ratios (AOR): 20.9; 95% confidence interval (CI): 2.2 - ∞). Environmental inspections, food/water testing, and a food trace-back investigation were inconclusive. Repeated cycles of cleaning were undertaken, including hydrogen peroxide fogging, however, transmission continued. After 7 months of investigation, environmental swabbing identified 106 isolates from kitchen surfaces and restaurant drains matching the outbreak profile. We found structural faults with the drainage system and hypothesised that a reservoir of bacteria in drain biofilm and underfloor flooded areas may have sustained this outbreak. Ineffective drain water-traps (U-bends) may have also contributed by allowing transmission of contaminated aerosols into the kitchen environment. These findings suggest that routine swabbing of sink drain points and inspection of drainage systems should be considered in future outbreak scenarios.

Re-evaluation of a 2014 multi-country European outbreak of Salmonella Enteritidis phage type 14b using recent epidemiological and molecular data.

A European multi-country outbreak of Salmonella Enteritidis phage type (PT) 14b occurred from March to November 2014 associated with the consumption of eggs. The outbreak involved more than 400 human cases from France, Luxembourg, Austria and the United Kingdom. In 2016-2017, it has been re-evaluated combining recent epidemiological results with latest molecular data. The outbreak was traced back to one large Bavarian egg producer with four distinct premises, three located in Bavaria, one in the Czech Republic. The outbreak isolates of S. Enteritidis PT 14b were grouped into three closely related clades by whole genome sequencing. Two of these clades could be referred to two Bavarian premises of the egg producer on the basis of epidemiological and molecular data, while epidemiological data presumably linked the third clade to another premises of the egg producer. Interestingly and in contrast to the situation in other European countries where several outbreaks were documented, all notified 91 laboratory-confirmed cases of S. Enteritidis PT 14b from Bavaria were sporadic, singular cases not belonging to any epidemiological outbreaks. In conclusion, as demonstrated here, the resolution of food-related outbreaks with such a high discriminatory power is rare in outbreak investigation.

Patchy promiscuity: machine learning applied to predict the host specificity of Salmonella enterica and Escherichia coli.

Salmonella enterica and Escherichia coli are bacterial species that colonize different animal hosts with sub-types that can cause life-threatening infections in humans. Source attribution of zoonoses is an important goal for infection control as is identification of isolates in reservoir hosts that represent a threat to human health. In this study, host specificity and zoonotic potential were predicted using machine learning in which Support Vector Machine (SVM) classifiers were built based on predicted proteins from whole genome sequences. Analysis of over 1000 S.enterica genomes allowed the correct prediction (67 -90 % accuracy) of the source host for S. Typhimurium isolates and the same classifier could then differentiate the source host for alternative serovars such as S. Dublin. A key finding from both phylogeny and SVM methods was that the majority of isolates were assigned to host-specific sub-clusters and had high host-specific SVM scores. Moreover, only a minor subset of isolates had high probability scores for multiple hosts, indicating generalists with genetic content that may facilitate transition between hosts. The same approach correctly identified human versus bovine E. coli isolates (83 % accuracy) and the potential of the classifier to predict a zoonotic threat was demonstrated using E. coli O157. This research indicates marked host restriction for both S. enterica and E. coli, with only limited isolate subsets exhibiting host promiscuity by gene content. Machine learning can be successfully applied to interrogate source attribution of bacterial isolates and has the capacity to predict zoonotic potential.

Salmonella Typhimurium gastroenteritis leading to chronic prosthetic vascular graft infection.

Introduction. It is estimated up to 6 % of prosthetic vascular grafts become infected. Staphylococcus aureus is predominant in early infection and coagulase-negative staphylococci are predominant in late infections. Enterobacteriaceae cause 14-40 % of prosthetic vascular graft infections. This is, to our knowledge the first reported case of Salmonella gastroenteritis causing chronic prosthetic vascular graft infection (PVGI). Case presentation. A 57 years old lady presented with signs and symptoms of prosthetic vascular graft infection. Three years earlier, she had undergone a prosthetic axillo-femoral bypass graft for critical limb ischaemia. The infected prosthetic vascular graft was removed and Salmonella Typhimurium was isolated on culture. In the intervening period, Salmonella Typhimurium was isolated from a faecal specimen, collected during an episode of acute gastroenteritis. Whole-genome sequencing (WGS) showed that the respective Salmonella Typhimurium isolates differed by only a single nucleotide polymorphism (SNP). Salmonella Typhimurium was not isolated on culture of a faecal specimen collected five days following cessation of antimicrobial therapy. Six months after removal of the prosthetic graft, the patient remains under follow-up for her peripheral vascular disease, which currently requires no further surgical intervention. Conclusion. This case has clear implications for the management of chronic PVGI. It is vital to collect high-quality surgical specimens for microbiological analysis and empirical choices of antibiotics are unlikely to cover all potential pathogens. It may also be prudent to enquire about a history of acute gastroenteritis when assessing patients presenting with chronic PVGI.

Recurrent seasonal outbreak of an emerging serotype of Shiga toxin-producing Escherichia coli (STEC O55:H7 Stx2a) in the south west of England, July 2014 to September 2015.

The first documented British outbreak of Shiga toxin-producing Escherichia coli (STEC) O55:H7 began in the county of Dorset, England, in July 2014. Since then, there have been a total of 31 cases of which 13 presented with haemolytic uraemic syndrome (HUS). The outbreak strain had Shiga toxin (Stx) subtype 2a associated with an elevated risk of HUS. This strain had not previously been isolated from humans or animals in England. The only epidemiological link was living in or having close links to two areas in Dorset. Extensive investigations included testing of animals and household pets. Control measures included extended screening, iterative interviewing and exclusion of cases and high risk contacts. Whole genome sequencing (WGS) confirmed that all the cases were infected with similar strains. A specific source could not be identified. The combination of epidemiological investigation and WGS indicated, however, that this outbreak was possibly caused by recurrent introductions from a local endemic zoonotic source, that a highly similar endemic reservoir appears to exist in the Republic of Ireland but has not been identified elsewhere, and that a subset of cases was associated with human-to-human transmission in a nursery.

Multinational outbreak of travel-related Salmonella Chester infections in Europe, summers 2014 and 2015.

Between 2014 and 2015, the European Centre for Disease Prevention and Control was informed of an increase in numbers of Salmonella enterica serotype Chester cases with travel to Morocco occurring in six European countries. Epidemiological and microbiological investigations were conducted. In addition to gathering information on the characteristics of cases from the different countries in 2014, the epidemiological investigation comprised a matched case-case study involving French patients with salmonellosis who travelled to Morocco that year. A univariate conditional logistic regression was performed to quantify associations. The microbiological study included a whole genome sequencing (WGS) analysis of clinical and non-human isolates of S. Chester of varied place and year of isolation. A total of 162 cases, mostly from France, followed by Belgium, the Netherlands, Spain, Denmark and Sweden were reported, including 86 (53%) women. The median age per country ranged from 3 to 38 years. Cases of S. Chester were more likely to have eaten in a restaurant and visited the coast of Morocco. The results of WGS showed five multilocus sequence types (ST), with 96 of 153 isolates analysed clustering into a tight group that corresponded to a novel ST, ST1954. Of these 96 isolates, 46 (48%) were derived from food or patients returning from Morocco and carried two types of plasmids containing either qnrS1 or qnrB19 genes. This European-wide outbreak associated with travel to Morocco was likely a multi-source outbreak with several food vehicles contaminated by multidrug-resistant S. Chester strains.

Two Linked Enteroinvasive Escherichia coli Outbreaks, Nottingham, UK, June 2014.

Enteroinvasive Escherichia coli (EIEC) outbreaks are uncommon in Europe. In June 2014, two EIEC outbreaks occurred in Nottingham, UK, within 2 days; outbreak A was linked to a takeaway restaurant and outbreak B to a wedding party. We conducted 2 analytical studies: a case-control study for outbreak A and a cohort study for outbreak B. We tested microbiological and environmental samples, including by using whole-genome sequencing. For both outbreaks combined, we identified 157 probable case-patients; 27 were laboratory-confirmed as EIEC O96:H19-positive. Combined epidemiologic, microbiological, and environmental findings implicated lettuce as the vehicle of infection in outbreak A, but the source of the organism remained unknown. Whole-genome sequencing identified the same organism in cases from both outbreaks, but no epidemiologic link was confirmed. These outbreaks highlight that EIEC has the capacity to cause large and severe gastrointestinal disease outbreaks and should be considered as a potential pathogen in foodborne outbreaks in Europe.

Detection of the plasmid-mediated mcr-1 gene conferring colistin resistance in human and food isolates of Salmonella enterica and Escherichia coli in England and Wales.

OBJECTIVES: In response to the first report of transmissible colistin resistance mediated by the mcr-1 gene in Escherichia coli and Klebsiella spp. from animals and humans in China, we sought to determine its presence in Enterobacteriaceae isolated in the UK. METHODS: The PHE archive of whole-genome sequences of isolates from surveillance collections, submissions to reference services and research projects was retrospectively analysed for the presence of mcr-1 using Genefinder. The genetic environment of the gene was also analysed. RESULTS: Rapid screening of the genomes of ∼24 000 Salmonella enterica, E. coli, Klebsiella spp., Enterobacter spp., Campylobacter spp. and Shigella spp. isolated from food or humans identified 15 mcr-1-positive isolates. These comprised: 10 human S. enterica isolates submitted between 2012 and 2015 (8 Salmonella Typhimurium, 1 Salmonella Paratyphi B var Java and 1 Salmonella Virchow) from 10 patients; 3 isolates of E. coli from 2 patients; and 2 isolates of Salmonella Paratyphi B var Java from poultry meat imported from the EU. The mcr-1 gene was located on diverse plasmids belonging to the IncHI2, IncI2 and IncX4 replicon types and its association with ISApl1 varied. Six mcr-1-positive S. enterica isolates were from patients who had recently travelled to Asia. CONCLUSIONS: Analysis of WGS data allowed rapid confirmation of the presence of the plasmid-mediated colistin resistance gene mcr-1 in diverse genetic environments and plasmids. It has been present in E. coli and Salmonella spp. harboured by humans in England and Wales since at least 2012.

Phylogenetic structure of European Salmonella Enteritidis outbreak correlates with national and international egg distribution network.

Outbreaks of Salmonella Enteritidis have long been associated with contaminated poultry and eggs. In the summer of 2014 a large multi-national outbreak of Salmonella Enteritidis phage type 14b occurred with over 350 cases reported in the United Kingdom, Germany, Austria, France and Luxembourg. Egg supply network investigation and microbiological sampling identified the source to be a Bavarian egg producer. As part of the international investigation into the outbreak, over 400 isolates were sequenced including isolates from cases, implicated UK premises and eggs from the suspected source producer. We were able to show a clear statistical correlation between the topology of the UK egg distribution network and the phylogenetic network of outbreak isolates. This correlation can most plausibly be explained by different parts of the egg distribution network being supplied by eggs solely from independent premises of the Bavarian egg producer (Company X). Microbiological sampling from the source premises, traceback information and information on the interventions carried out at the egg production premises all supported this conclusion. The level of insight into the outbreak epidemiology provided by whole-genome sequencing (WGS) would not have been possible using traditional microbial typing methods.

Whole Genome Sequencing demonstrates that Geographic Variation of Escherichia coli O157 Genotypes Dominates Host Association.

Genetic variation in an infectious disease pathogen can be driven by ecological niche dissimilarities arising from different host species and different geographical locations. Whole genome sequencing was used to compare E. coli O157 isolates from host reservoirs (cattle and sheep) from Scotland and to compare genetic variation of isolates (human, animal, environmental/food) obtained from Scotland, New Zealand, Netherlands, Canada and the USA. Nei's genetic distance calculated from core genome single nucleotide polymorphisms (SNPs) demonstrated that the animal isolates were from the same population. Investigation of the Shiga toxin bacteriophage and their insertion sites (SBI typing) revealed that cattle and sheep isolates had statistically indistinguishable rarefaction profiles, diversity and genotypes. In contrast, isolates from different countries exhibited significant differences in Nei's genetic distance and SBI typing. Hence, after successful international transmission, which has occurred on multiple occasions, local genetic variation occurs, resulting in a global patchwork of continental and trans-continental phylogeographic clades. These findings are important for three reasons: first, understanding transmission and evolution of infectious diseases associated with multiple host reservoirs and multi-geographic locations; second, highlighting the relevance of the sheep reservoir when considering farm based interventions; and third, improving our understanding of why human disease incidence varies across the world.

Analysis of whole genome sequencing for the Escherichia coli O157:H7 typing phages.

BACKGROUND: Shiga toxin producing Escherichia coli O157 can cause severe bloody diarrhea and haemolytic uraemic syndrome. Phage typing of E. coli O157 facilitates public health surveillance and outbreak investigations, certain phage types are more likely to occupy specific niches and are associated with specific age groups and disease severity. The aim of this study was to analyse the genome sequences of 16 (fourteen T4 and two T7) E. coli O157 typing phages and to determine the genes responsible for the subtle differences in phage type profiles. RESULTS: The typing phages were sequenced using paired-end Illumina sequencing at The Genome Analysis Centre and the Animal Health and Veterinary Laboratories Agency and bioinformatics programs including Velvet, Brig and Easyfig were used to analyse them. A two-way Euclidian cluster analysis highlighted the associations between groups of phage types and typing phages. The analysis showed that the T7 typing phages (9 and 10) differed by only three genes and that the T4 typing phages formed three distinct groups of similar genomic sequences: Group 1 (1, 8, 11, 12 and 15, 16), Group 2 (3, 6, 7 and 13) and Group 3 (2, 4, 5 and 14). The E. coli O157 phage typing scheme exhibited a significantly modular network linked to the genetic similarity of each group showing that these groups are specialised to infect a subset of phage types. CONCLUSION: Sequencing the typing phage has enabled us to identify the variable genes within each group and to determine how this corresponds to changes in phage type.

Insight into Shiga toxin genes encoded by Escherichia coli O157 from whole genome sequencing.

The ability of Shiga toxin-producing Escherichia coli (STEC) to cause severe illness in humans is determined by multiple host factors and bacterial characteristics, including Shiga toxin (Stx) subtype. Given the link between Stx2a subtype and disease severity, we sought to identify the stx subtypes present in whole genome sequences (WGS) of 444 isolates of STEC O157. Difficulties in assembling the stx genes in some strains were overcome by using two complementary bioinformatics methods: mapping and de novo assembly. We compared the WGS analysis with the results obtained using a PCR approach and investigated the diversity within and between the subtypes. All strains of STEC O157 in this study had stx1a, stx2a or stx2c or a combination of these three genes. There was over 99% (442/444) concordance between PCR and WGS. When common source strains were excluded, 236/349 strains of STEC O157 had multiple copies of different Stx subtypes and 54 had multiple copies of the same Stx subtype. Of those strains harbouring multiple copies of the same Stx subtype, 33 had variants between the alleles while 21 had identical copies. Strains harbouring Stx2a only were most commonly found to have multiple alleles of the same subtype (42%). Both the PCR and WGS approach to stx subtyping provided a good level of sensitivity and specificity. In addition, the WGS data also showed there were a significant proportion of strains harbouring multiple alleles of the same Stx subtype associated with clinical disease in England.

Whole Genome Sequencing for the Retrospective Investigation of an Outbreak of Salmonella Typhimurium DT 8.

BACKGROUND: Salmonella enterica serovar Typhimurium DT8 is uncommon within the European Union. An increase in this phage type was reported in the summer of 2013 in the States of Jersey. METHODS: A total of 21 human cases with this phage type were microbiologically confirmed. Salmonella isolates from mayonnaise made using raw eggs were also confirmed as being Salmonella Typhimurium DT8. The epidemiological investigations strongly supported a link between mayonnaise consumption and illness. Whole genome sequencing (WGS) was used to retrospectively investigate this outbreak with a view to assess the similarity between the suspect food and the human isolates and to characterise a known point source outbreak to assist in development of algorithms for outbreak detection. RESULTS: Sequence data showed that the outbreak associated isolates, including the food isolates, formed a tightly clustered monophyletic group, with a maximum pairwise distance of 3 single nucleotide polymorphisms. CONCLUSIONS: WGS data is useful in confirming the causative agent of outbreaks where food and clinical isolates are available. This dataset, comprising a known outbreak, will be useful in the development of automatic algorithms for outbreak detection.

MinION nanopore sequencing identifies the position and structure of a bacterial antibiotic resistance island.

Short-read, high-throughput sequencing technology cannot identify the chromosomal position of repetitive insertion sequences that typically flank horizontally acquired genes such as bacterial virulence genes and antibiotic resistance genes. The MinION nanopore sequencer can produce long sequencing reads on a device similar in size to a USB memory stick. Here we apply a MinION sequencer to resolve the structure and chromosomal insertion site of a composite antibiotic resistance island in Salmonella Typhi Haplotype 58. Nanopore sequencing data from a single 18-h run was used to create a scaffold for an assembly generated from short-read Illumina data. Our results demonstrate the potential of the MinION device in clinical laboratories to fully characterize the epidemic spread of bacterial pathogens.

Enteroaggregative Escherichia coli have evolved independently as distinct complexes within the E. coli population with varying ability to cause disease.

Enteroaggregative E. coli (EAEC) is an established diarrhoeagenic pathotype. The association with virulence gene content and ability to cause disease has been studied but little is known about the population structure of EAEC and how this pathotype evolved. Analysis by Multi Locus Sequence Typing of 564 EAEC isolates from cases and controls in Bangladesh, Nigeria and the UK spanning the past 29 years, revealed multiple successful lineages of EAEC. The population structure of EAEC indicates some clusters are statistically associated with disease or carriage, further highlighting the heterogeneous nature of this group of organisms. Different clusters have evolved independently as a result of both mutational and recombination events; the EAEC phenotype is distributed throughout the population of E. coli.