Development of food safety risk assessment tools based on molecular typing and WGS of Campylobacter jejuni genome

Abstract

The ‘learning‐by‐doing’ EU‐FORA fellowship programme in the development of risk assessment tools based on molecular typing and WGS of Campylobacter jejuni genome was structured into two main activities: the primary one focused on training on risk assessment methodology and the secondary one in starting and enhancing the cooperation between the hosting and home organisations, or other joint activities. The primary activities had three subsequent work packages (WPs): WP1 data organisation, WP2 cluster and association analyses, and WP3 development of risk assessment models. The secondary activities have branched into one workshop and the initiation of a cooperation programme between the hosting and home organisations. In the last quarter, the fellow had contributed to the characterisation of some pathogens in possible response to a changing climate, part of the CLEFSA project. The fellow attended various forms of training: online and on‐site courses, and also participated at several conferences and meetings for improving his knowledge and skills, contributing to performing the Campylobacter risk assessment and source attribution.

1. Introduction

1.1. About EU‐FORA

For building the European Union risk assessment capacity and knowledge community, the European Food Safety Authority (EFSA) initiated the European Food Risk Assessment Fellowship Programme (EU‐FORA), as part of the EFSA Strategic Objective. The programme is focused on training in risk assessment and in intensifying exchange and cooperation between different organisations and EFSA. The training is based on ‘learning by doing’ risk assessment methodologies and practices by involving the fellow in one project linked with food safety from the hosting site, where risk assessment is a significant part. The Istituto Zooprofilattico Sperimentale dell’ Abruzzo e del Molise ‘G. Caporale’ (IZSAM) initiated many research projects in this field through the Italian National Reference Centre for Veterinary Epidemiology, Programming, Information and Risk Analysis.

1.2. General framework

The Campylobacter spp. are worldwide distributed bacteria and represent one important microbiological hazard linked with food‐borne zoonosis. The gastroenteritis caused by C. jejuni and C. coli consists of 6 days watery or bloody diarrhoea, self‐limiting disease in the majority of the cases, with or without associated fever, weight loss, cramps and headache (Man, 2011; MSD, 2019). Furthermore, C. jejuni is associated with a range of other gastrointestinal and extra‐gastrointestinal infectious conditions including bacteremia and sepsis, also it may lead to autoimmune conditions known as Guillain–Barré syndrome (GBS), Miller Fisher syndrome, irritable bowel syndrome and Bell's palsy (unilateral facial paralysis), and colorectal cancer (Kaakoush et al., 2015; MSD, 2019). Because the precise role of Campylobacter species in the development of these clinical conditions is unknown and the highest prevalence of the gastrointestinal reported conditions caused by this genus, more accurate methodologies in their surveillance and monitoring are necessary. For these reasons the surveillance of Campylobacter spp. is part of the EFSA strategy. The legal framework for its monitoring is Regulation (EC) 2160/2003 (European Union, 2003aa) on the control of Salmonella and other specified food‐borne zoonotic agents and Directive 2003/99/EC on the monitoring of zoonosis and zoonotic agents, however, reporting Campylobacter infection is not mandatory in all countries (European Union, 2003bb).

Considering the concern in public health of the Campylobacter spp. contaminations along the production chain of some food products, the IZSAM is currently involved in several epidemiology studies. The following IZSAM's units are involved: the Italian National Reference Centre for Veterinary Epidemiology, Programming, Information and Risk Analysis (COVEPI), the Italian National Reference Centre for Whole Genome Sequencing of microbial pathogens: database and bioinformatics analysis (GENPAT), and the Italian National Reference Laboratory for Campylobacter (NRL).

2. Description of work programme

2.1. Aims

The first objective of the work programme was ‘learning‐by‐doing’ of the fellow in the food safety risk assessment methodology, including collecting, normalisation, and analysis of the data and involving in developing and validation of a set of risk assessment epidemiological tools based on molecular typing and WGS of Campylobacter jejuni genomes.

The work programme had three subsequent work packages (WP), focussed on C. jejuni:

• WP1. Data organisation: focuses on reviewing the literature, collecting the raw data available, analysing, normalising and organising the data for subsequent epidemiological analyses.

• WP2. Cluster and association analyses: with the purpose to understanding and gaining practical skills in bioinformatics, becoming familiar with tools used in bioinformatics, being trained by doing various statistical methods and approaches for data analysis.

• WP3. Development of risk assessment models: through involvement in the development and validation of a set of risk assessment model considering the main C. jejuni genotypes and estimating their contribution in the whole exposure of consumers from Italy.

The second objective was to enhance relationships among home and hosting organisation. The IZSAM is FAO Reference Centre for Veterinary Epidemiology and OIE Collaborating Centre for Veterinary Training, Epidemiology, Food Safety and Animal Welfare, and finding a solution for agreement and cooperation between organisations was considered.

The fellow has been part of the COVEPI's team involved in the activities of this working programme. In particular, two senior epidemiologists and a statistician have been part of the team, which worked closely with the bioinformaticians of the GENPAT, and the personnel of the NRL for Campylobacter. The two senior epidemiologists of the COVEPI's team have acted as mentors for the fellow, assisting him on a daily basis during all the activities carried out. A coordination meeting among the fellow, mentor and supervisor has been held regularly, on a weekly basis, to verify the work done and discuss the implications of the results obtained and plan subsequent activities. The fellow specifically worked on the analysis of the data generated by the molecular typing of C. jejuni, and in developing, testing and validation of different tools for the identification of spatiotemporal clusters of epidemiological relevance. The fellow had followed the whole data production and processing process, from the sequencing activities, carried out at laboratory to the bioinformatics analyses and statistical analyses performed on sequence data. In addition, the fellow had benefited greatly by the participation of IZSAM to the activities of the COHESIVE project, with his involvement in selected project meetings.

2.2. Activities/methods

2.2.1. WP1. Data organisation

In the beginning, the advanced literature search has been conducted regarding the EFSA and international guidance in risk assessment and source attribution based on whole genome sequencing and antimicrobial resistance (AMR), to learn and understand better what is the actual scientific level in the field and what are the best ways to approach the study (and not limited to that, including also the scientific literature in the field) (EFSA BIOHAZ Panel, 2013).

The purpose of WP1 was to obtain the Campylobacter database containing the molecular typing results and related epidemiologically relevant metadata, ready to be analysed. The large part of this WP it was made in the first step at the Italian National Reference Laboratory (NRL) for Campylobacter. The protocol to Campylobacter spp. surveillance is routinely performed by Italian NRL's and is made by cultivation and identification according to EN ISO 10272 part 1 and 2 method (ISO, 2006: Parts 1 and 2). The genotyping characterisation of the isolates are made through species typing by molecularly confirmation by multiplex polymerase chain reaction (PCR) (Wang et al., 2002), flaA‐SVR sequencing (Nachamkin et al., 1993), pulsed‐field gel electrophoresis (PFGE) (Institue of Environmental Science and Research, 2013) and BioNumerics software version 7.6 (Applied Maths, 2019), in silico multilocus sequence typing (MLST) (Dingle et al., 2008; Jolley et al., 2018; Seemann, 2019), and since 2017, C. jejuni isolates are progressively submitted to Core Genome MLST (cgMLST) for a better discriminatory analysis (O'Mahony et al., 2011; Llarena et al., 2017). Part of these sequencing activities is carried out under various national and international projects. The AMR phenotype characterisation is made by testing of susceptibility to seven antimicrobials with a micro‐broth dilution method using the ‘Sensititre’ automated system (TREK Diagnostic Systems, Biomedical Service, Italy) (Kittl et al., 2013; EFSA, 2019).

To date about 3,000 isolates of Campylobacter spp., respectively, collected in entire Italy, are available in the IZSAM's strains collection. In this step, the fellow was actively involved in some specific analytical technique linked with cultivation, genotypic and phenotypic characterisation of Campylobacter, having the opportunity to develop and improve his laboratory skills and get acquainted with the National Veterinary Information System (https://www.vetinfo.sanita.it/), managed by IZSAM. During this WP, the fellow had learned and understood the entire laboratory workflow (Figure 1) until obtaining the raw data and their recording into the Italian information system. At the same time, the fellow became familiar with the international Campylobacter MLST database (https://pubmlst.org/) (Jolley et al., 2018), and used it for comparison and normalisation of the data. IZSAM collects and registers a well‐defined set of standardised data for each sample tested in its laboratories. In addition, several samples are collected in the framework of national control plans and all related data are registered into the National Veterinary Information System. All these factors allow IZSAM to retrieve relevant epidemiological data for all tested samples. These epidemiological metadata are fundamental for a correct interpretation of the microbiological results, including the outcomes of molecular typing and phenotyping (ex: antimicrobial resistance). A second step has been done in COVEPI, where a detailed data analysis plan has been developed, including the description of the dataset to be retrieved, the type of data quality checks to be performed and the format of the resulting validated databases. After extracting the raw data, the fellow retrieved, verified, normalised and organised the data for subsequent epidemiological analyses. These data were used in source attribution for human illness through microbial subtyping. (EFSA, 2008, 2019)

Figure 1.

Laboratory flow chart for Campylobacter spp. analysis

2.2.2. WP2. Cluster and association analyses

WP2 was focused on obtaining the results from the analyses of molecular typing data and identification of main ‘epi‐clusters’ of C. jejuni by analysing the genotypes and phenotypes data as well as epidemiological metadata. In the first step, the fellow had the opportunity to work in the Italian National Reference Centre for Whole Genome Sequencing of microbial pathogens (GENPAT). The bioinformaticians of the GENPAT supported the manipulation and analysis of sequencing data and introduced the fellow in bioinformatics and familiarised him to the bioinformatics software (Figure 3). For comparative analysis based on cgMLST data, the fellow learned and used the minimum spanning tree with GrapeTree software (Figure 2) (Zhou et al., 2018). The second step had been in COVEPI's Statistics and GIS Unit where the fellow was trained in using QGIS geographic information system software. During the third step, in the COVEPI, the fellow analysed the data for the purpose of verifying statistically significant genetic clusters of Campylobacter spp. (C. jejuni) associated with: the particular species, farm types, food production products; specific phenotypic characteristics, like AMR patterns; or specific spatiotemporal patterns and persistence in specific groups of farms. Working with a big amount of data was challenging, prompting to consider to use some business intelligence software like MicroStrategy.

Figure 3.

The cgMLST data flow chart for Campylobacter spp.

Figure 2.

Bioinformatics flow chart for Campylobacter spp. analysis

2.2.3. WP3. Development of risk assessment models

During WP3, in COVEPI unit, different source attribution methods had been used for finding and characterisation of the main C. jejuni genotypes and estimating their contribution the whole exposure of Italian consumers. The development and validation of a risk assessment model considering the genotypes and phenotypes were considered. In this step, the fellow analysed the microbial genotypic and phenotypic characteristics of the isolates, to find the common fingerprint and to define the ‘epi‐cluster’. Within the team, different biostatistical methodologies were tested.

Another approach has been attempted in order to define a particular methodology designated to assess the sources of uncertainties in source attribution models used in Campylobacter microbial risk assessment (MRA). The considered methodologies were: failure mode effect analysis (FMEA), fault tree analysis (FTA) and key process indicators (KPI)/quality indicators (QIs). Before starting the assessment, the actors and the stages needed to be defined (pre‐preanalytical, preanalytical, analytical, postanalytical and post‐postanalytical).

2.2.4. Secondary activities

Secondary activities are referred about agreements and cooperation between hosting sites and organisations of origin, cooperation between different organisations and also to other activities not mentioned before, in which the fellow played an active role.

2.2.4.1. Agreements and cooperation

In view of recent epidemics in Europe, taking advantage of this fellowship programme during this period, the foundations have been laid for establishing and strengthening the collaboration between the national reference centres in Italy and Romania, from the IZSAM, Teramo, and the Institute for Diagnosis and Animal Health (IDAH), Bucharest. The first step was to have 2 days ‘Animal health risk assessment and vector‐borne diseases’ workshop during the 2–3 April 2019 for the specialists from IDAH in Bucharest, under scientific coordination of the fellow and participation of the tutors from IZSAM, Paolo Calistri, and Federica Monaco, and from Agricultural Research Council – Onderstepoort Veterinary Research (ARC‐OVR), Gert Venter. The second step was the assessment of the needs for professional training in epidemiology for Romanian specialists, for the purpose to find the optimum solution to improve and increase the animal health risk assessment capacity. Depending on the size of the training needs, consideration has been given to the temporary training of a limited number of specialists at the IZSAM, or the initiation of a twinning project, if the needs are more complex. This process is in progress.

In the same context, the European BioSafety Association (EBSA), during their 22nd annual conference at Bucharest, from the total of 42, offered sponsored participation for 24 specialists from Romania including from IDAH and ANSVSA, for participation at the preconference courses (8) in the field of biosafety and biosecurity, and also for participation at the conference (6). Considering the importance of the topics in this region, EBSA increased more than fourth times the numbers of sponsored participation for the persons from Balkan region (Romania, Bulgaria, Republic of Moldova, Albania, Croatia, Ukraine, Georgia, and Greece) for this year. After this successful experience, different types of collaboration between organisations and their members have started to be considered.

In the last quarter of the EU‐FORA programme, the fellow had contributed to the characterisation of some pathogens (like Campylobacter jejuni) in possible response to a changing climate, in terms of possible increase in exposure or pathogenicity under a specific climate change scenario, part of the CLEFSA project, coordinated by Angelo Maggiore (EFSA).

2.2.4.2. Additional activities

During 13–15 November 2018, the fellow participated at the Romanian National Sanitary Veterinary and Food Safety Authority (ANSVSA) meeting in Baia‐Mare, Romania, for the presentation of the EU‐FORA fellowship programme and introduction in risk management, and basic concepts in risk assessment, to the specialists from the national laboratory network.

The fellow participated in the workshop ‘Accounting for uncertainty in data‐poor scenarios: Case studies on risk analysis in food safety’ and at the International Conference on Uncertainty in Risk Analysis, 20–22 February 2019, Berlin at the German Federal Institute for Risk Assessment (BfR), with ‘Uncertainty assessment in Campylobacter spp. source attribution models: some qualitative approaches’ poster presentation at ‘Methods for uncertainty analysis’ thematic area.

The fellow participated at the 22nd annual international conference ‘Burning topics in Biosafety’ of the EBSA at Bucharest, Romania, in 2–5 April 2019, where together with the experts Paolo Calistri (IZSAM) and Uwe Mueller‐Doblies (MSD), moderated the break‐out ‘Biosafety and biosecurity in the field in case of an emergency’. Also, the fellow presented two posters: ‘The exposure to Campylobacter spp. of the food industry workers: a short overview’ and ‘EFSA EU‐FORA – The European Food Risk Assessment Fellowship Programme’.

Like invited guest and collaborator, the fellow participated at the ‘Transylvanian Experimental Neuroscience Summer School – TENSS 2019’ during 15–16 June 2019, Pike Lake, Romania, organised by the Transylvanian Institute of Neuroscience (TINS).

3. Conclusions

3.1. Risk assessment in C. jejuni

The EFSA EU‐FORA ‘learning by doing’ programme is among the few from Europe which is also addressed to mid‐career scientists and open to those who do not necessarily come from the academic field, being a real fortune for people from East European countries. This was a great opportunity for the fellow to consolidate his specialised knowledge and skills in food safety and veterinary epidemiology and public health, by working in a prestigious international and national reference centre. He gained experience by participating in the dedicated program and other activities within the host organisation, better understanding the complex workflow in the specific risk assessment and, at the same time, learned how to investigate an outbreak epidemic, its strengths and weaknesses.

This training offered the opportunity to work both independently and in a team, to integrate the multidisciplinary and evidence‐based veterinary medicine approach into assessing the risk of Campylobacter and assigning the source in particular.

Applying microbial subtyping methodology in Campylobacter source attribution, some characteristics of its population in Italy has been identified. The complexity and the high volume of data provided by cgMLST, at this time, in the absence of a common database, rendered somehow difficult the source attribution based just on this information, anyway, the evidence indicates the existence within C. jejuni population of one ‘relatively stable’ cluster and one ‘very dynamic’. Analysing the AMR fingerprint data of the ‘relative stabile’ C. jejuni population, considering the characteristics of the antimicrobial tested, even in the absence of the direct evidence, has supposed that the population had been selected from strains located from another site than the intestinal tract of the animals. The data gathered were not obtained on the basis of a well‐defined sampling program, however, the complexity of the information obtained and analysed allows the initiation of a description of the C. jejuni population from Italy, that will represent the foundation for building the common database and harmonised methodology for sub‐typing, analysis and storage the data and, in the future, the development of linkage mechanisms (EFSA BIOHAZ Panel, 2013, 2014).

After applying in microbial risk assessment in poultry of some basic cause‐effect and effect‐cause models (like FMEA respectively FTA) and key process indicators methodologies, a potential scoring system and KPI have been designed and proposed (Table 1) (Wikipedia, the free encyclopedia, 2019a,b) The results were used to draw the flow chart in source attribution in poultry (Figure 4).

Table 1.

The list of some key performance indicators (KPI)/quality indicators (QIs) tailored for risk assessment

Stage	Weighting field	KPI/QIs
The KPI with priority 1
Pre‐preanalytical	Weighting the average of the event	a)Number of events, incident or accident/1,000 units of goods b)Number of events, incident or accident/1,000 units of time c)Number of positive Campylobacter spp. samples/1,000 units of goods d)Number of positive Campylobacter spp. samples/units of time e)Number of positive Campylobacter spp. CFU/g per 1,000 units of goods number of positive Campylobacter spp. CFU/g per units of time
Pre‐preanalytical	Weighting the average of the manufacturing process	a)Number of goods/units of time b)Number of servings/unit of goods c)Number of personnel/organisation d)Number of personnel necessary/1,000 units of goods e)Number of personnel necessary/units of time
Pre‐preanalytical	Weighting the average of the transport process	a)Number of goods/units of transport b)Number of units of transport/unit of time c)Number of goods transported/units of time

CFU: colony forming unit.

The KPI was defined with priority 1, compulsory; 2, important; 3, proposed; 4, valuable.

Figure 4.

Fault tree analysis diagram in poultry carcass contamination, considering the source of Campylobacter spp. from poultry

During this programme, the fellow had gained skills in using different software like GrapeTree, QGIS, MicroStrategy and had been initiated in using R.

3.2. Building cooperation

The future agreement and collaboration between the IZSAM, Teramo, and the IDAH, Bucharest, represents a priority especially for the Romanian part. In this context, several fields for collaboration have been designed, like laboratory activity and epidemiology, public health, and risk analysis.

Many fellows from the 2nd series of the EU‐ FORA programme had worked together at the poster for the EBSA conference, named ‘The exposure to Campylobacter spp. of the food industry workers: a short overview’. The greatest gain of this project is represented by the established human relationships and networking. Starting with relationships with the EFSA coordinators experts, continuing with program coordinators, tutors, experts, and personnel from the hosting sites, and last but not least with the fellow colleagues, this series adds a brick at the building of the future of the EU.

Abbreviations

AMR

antimicrobial resistance

ANSVSA

Romanian National Sanitary Veterinary and Food Safety Authority

ARC‐OVR

Agricultural Research Council – Onderstepoort Veterinary Research, South Africa

BfR

German Federal Institute for Risk Assessment

CFU

colony forming unit

cgMLST

Core Genome MultiLocus Sequence Typing

CLEFSA

Climate change as a driver of emerging risks for food and feed safety, plant, animal health and nutritional quality (CLEFSA)’ project conducted by the EFSA Scientific Committee ‐ Emerging Risks Unit (SCER)

COHESIVE

Cohesive –One Health Structure In Europe‐ is a 3‐year project, which aims to develop sustainable One Health approaches with respect to signalling, assessing and controlling zoonoses at the national level within EU countries and across borders

COVEPI

Italian National Reference Centre for Veterinary Epidemiology, Programming, Information and Risk Analysis (IZSAM)

EBSA

European BioSafety Association

EU‐FORA

The European Food Risk Assessment Fellowship Programme

FAO

The Food and Agriculture Organization of the United Nations

flaA‐SVR

The DNA sequence of the flagellin A short variable region

FMEA

failure mode effect analysis

FTA

fault tree analysis

GENPAT

Italian National Reference Centre for Whole Genome Sequencing of microbial pathogens: database and bioinformatics analysis (IZSAM)

IDAH

Institute for Diagnosis and Animal Health, Bucharest, Romania

IZSAM

Istituto Zooprofilattico Sperimentale dell’ Abruzzo e del Molise “G. Caporale”, Teramo, Italy

KPI

key process indicators

MLST

multilocus sequence typing

MRA

microbial risk assessment

MSD

Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc. Kenilworth, NJ, USA

NRL

National Reference Laboratory

OIE

The World Organisation for Animal Health

PCR

polymerase chain reaction

PFGE

pulsed‐field gel electrophoresis

QIs

quality indicators

TEENS‐2019

Transylvanian Experimental Neuroscience Summer School –2019

TINS

The Transylvanian Institute of Neuroscience, Cluj‐Napoca, Romania

VBD

vector‐borne diseases

WGS

whole genome sequencing

WP

work package

Appendix A – The on‐site courses and training where fellow attended

1.

Subject	Organisation	Location	Period	Time	Tutor's
Internal training on risk assessment methods	IZSAM	Teramo, Italy	29.10.2018	6 h	Paolo Calistri
Introduction in bioinformatics	IZSAM	Teramo, Italy	8.11.2018	2 h	Adriano Di Pasquale Antonio Rinaldi
The use of GIS and QGIS software	IZSAM	Teramo, Italy	13.3–13.4.2019	16 h	Susanna Tora
Microstrategy Dashboarding Data with Dossiers	Microstrategy	Milano, Italy	11.3.2017	8 h	Stefano Sartorio
Microstrategy Dashboarding Data with Dossiers	Microstrategy	Roma, Italy	6.5.2017	8 h	Stefano Sartorio
Microstrategy Advanced Reporting	Microstrategy	Roma, Italy	7.5.2017	8 h	Stefano Sartorio
Microstrategy Enterprise Mobility	Microstrategy	Roma, Italy	8.5.2017	8 h	Stefano Sartorio
Microstrategy Enterprise Applications	Microstrategy	Roma, Italy	9.5.2017	8 h	Stefano Sartorio
Transylvanian Experimental Neuroscience Summer School – TENSS 2019 (advanced modelling of biological systems, basic and advanced concepts in signal processing, statistical methods for the evaluation of dynamical systems, principal component analysis, machine learning)	TINS	Pike Lake, Romania	15–16 June 2019	6 h	Raul C. Muresan Vasile V.Moca Christian Machens Fede Carnevale

Appendix B – The online courses and training where fellow attended

1.

Subject	Organisation	Period	Time	Tutor's
The FoodEx2 classification system and guidance on its harmonised use	EFSA	26 September 2018	1 h	Sofia Ioannidou Laura Kirwan Alban Shahaj
The FoodEx2 classification system and guidance on its harmonised use	EFSA	3 October 2018	1 h	Sofia Ioannidou Laura Kirwan Alban Shahaj
How to report surveillance data on Transmissible Spongiform Encephalopathies using the EFSA tool	EFSA	21 January 2019	1 h
Learn more about the risk assessment of phthalates used in plastic food contact materials	EFSA	15 March 2019	1 h
EFSA's new dedicated support to SMEs	EFSA	24 May 2019	1/2 h	Remigio Marano Patricia Romero
EPI‐interactive webinar: Introduction to R Shiny	EPI‐Interactive	30 May 2019	1 h	Uli Muellner
IHU BioSecurity Free Webinar	International Hellenic University	15 May 2019	2 h	Gijsbert van Willigen Patrick Rüdelsheim
Medical School Pathology Courses	Dr. Minarcik's Online Medical School Pathology Course	1.9.2018–31.5.2019	24 h	John R. Minarcik
Illness Outbreaks linked to Enteric Zoonoses and the Interconnectedness of Human and Animal Health	Pet Poison Helpline	24 April 2019	1 h	Megin Nichols

Appendix C – The summary agenda of ‘Animal health risk assessment and vector‐borne diseases’ workshop at Bucharest

1.

Subject	Tutors
The Culicoides diseases in an endemic area	G. Venter
The vector collection methods	G. Venter
Workgroup schedule	G. Venter/A.I. Ardelean
Epidemiological surveillance and risk factors	P. Calistri
Introduction to risk assessment in animal health	P. Calistri
Lumpy skin disease: epidemiology and control aspects	F. Monaco/P. Calistri
Surveillance of West Nile disease in Italy: example of an integrated One Health approach	F. Monaco/P. Calistri
Geographical information systems: tools for VBD surveillance and control	P. Calistri

VBD: vector‐borne disease.

Annex A – Poster Uncertainty assessment in Campylobacter spp. source attribution models: some qualitative approaches

1.

Annex B – Poster EFSA EU‐FORA – The European Food Risk Assessment Fellowship Programme

1.

Annex C – Poster The exposure to Campylobacter spp. of the food industry workers: a short overview

1.

Suggested citation: Istituto Zooprofilattico Sperimentale dell’ Abruzzo e del Molise “G. Caporale”, Teramo, Italy , Ardelean AI, Calistri P, Giovannini A, Garofolo G, Di Pasquale A, Conte A and Morelli D, 2019. Development of food safety risk assessment tools based on molecular typing and WGS of Campylobacter jejuni genome. EFSA Journal 2019;17(S2):e170903, 18 pp. 10.2903/j.efsa.2019.e170903