EFSA's WGS System: Enhancing Food Safety in the EU by Connecting Genomic Data

Whole genome sequencing (WGS) has revolutionized our understanding of and ability to combat foodborne illnesses by transforming routine surveillance, outbreak detection, outbreak response, and source identification. WGS plays a crucial role in enhancing surveillance efforts, enabling a more comprehensive understanding of the complex dynamics of foodborne diseases. By harnessing the power of WGS, outbreaks can be rapidly and accurately detected, facilitating prompt and targeted response measures.¹ Furthermore, WGS enables the identification of foodborne illness sources, aiding in the prevention of further contamination and spread. This ultimately enhances the safety of the food chain while alleviating the economic burden associated with foodborne illness outbreaks.²

Challenges in Implementing Genomic Sequencing for Food Safety

Despite the evident impact that WGS is making in the field of foodborne pathogen surveillance and outbreak response, its implementation still faces numerous challenges, particularly within the European context.^1,3 The establishment of a comprehensive surveillance program that harnesses the benefits of WGS requires a coordinated approach involving multiple stakeholders at multiple levels within the framework of One Health. These stakeholders include public health agencies, food safety organizations, veterinary bodies, and regulatory authorities. However, achieving seamless coordination is hindered by disparities in organizational structures, cultural and political factors at local or national levels, as well as technical and operational challenges. Existing legislation, differing priorities, and resistance to change can impede the potential beneficial impact of WGS in enhancing public health and food safety response during crises, where timely data production and sharing are critical for early outbreak detection and effective outbreak response. Also, legal and ethical hurdles regarding data ownership and responsible data usage can lead to the reluctance to share data across sectors and between nations.

Economic factors influence the readiness of certain countries to adopt routine WGS methodologies. Indeed, the costs associated with establishing and maintaining sequencing and computational technologies can create barriers to implementation. Capacity development at the country level is, therefore, essential to ensure the production of high-quality and reproducible data. On the technical front, challenges arise concerning data protection, storage, and sharing of WGS data and associated contextual data due to infrastructure and technical requirements. Scientific considerations also play a crucial role in the effective utilization of WGS data. With the growing diversity of sequencing platforms and bioinformatics tools available, selecting the most suitable platform and analytical pipelines becomes critical. Ensuring data quality and harmonization is, therefore, paramount to facilitate meaningful comparisons and reliable analyses.¹

Genomic Data Sharing: Collaborative Approaches for Safer Food

Numerous notable initiatives worldwide, including GenomeTrkr,⁴ PulseNet International,⁵, and Global Initiative on Sharing All Influenza Data (GISAID),⁶ have adopted diverse data-sharing models to effectively utilize WGS data.

These data-sharing models differ in terms of their approach to openness and accessibility, reflecting unique priorities and requirements within their respective domains. Some models prioritize unrestricted access to genomic data, empowering researchers and the general public to freely explore and utilize the information. Conversely, other models take a more controlled approach, sharing data within trusted networks of stakeholders to ensure responsible usage.

Moreover, these initiatives have also made distinct decisions regarding the analytical components, which directly impact the type of data that is stored and shared. This includes choices such as sharing either raw sequencing reads or transformed data. Additionally, variations exist in the collection and sharing of contextual data, as well as in the application of relevant standards. These factors influence the interoperability of data and contribute to the overall effectiveness of the models.

It is evident that a one-size-fits-all model does not exist, as different settings have varying requirements, necessitating tailored approaches.

Facilitating Data Exchange for Food Safety in EU: The EFSA One Health WGS System

The European Food Safety Authority (EFSA) has placed a high priority on establishing a robust infrastructure to securely and efficiently store and share WGS data and associated contextual data in the European Union (EU) and European Economic Area (EEA). EFSA's goal is to foster collaboration and knowledge exchange among risk assessors and managers at the EU level while ensuring responsible use of shared information. In the multinational European context, selecting an appropriate data-sharing model that ensures a responsible chain of custody and expedites risk assessment for the protection of human health is crucial, particularly for food and veterinary data. This model should allow data owners (i.e., EU and EEA countries) to retain ownership of their valuable information while sharing it with trusted entities, emphasizing data integrity and responsible data usage. By adopting this fit-for-purpose model, data can be shared effectively while maintaining necessary safeguards, prioritizing operational needs, and enabling swift actions to protect public health.⁷

Considering these factors, EFSA put into operation the EFSA One Health WGS system in July 2022, following a mandate sent by the European Commission to both EFSA and the European Center for Disease Prevention and Control (ECDC). The goal of this mandate was to create two interconnected systems among the two agencies to facilitate the efficient collection, analysis, and sharing of WGS data and contextual data at the European level.⁸ These systems operate within their designated domains, encompassing the food and veterinary sector (EFSA) and the public health sector (ECDC). Through adherence to specific rules, these systems enable the rapid detection of multi-country foodborne outbreaks by facilitating data exchange and collaboration across countries.

The EFSA One Health WGS System works in a close environment. Its access is restricted and not freely available to the general public. Only individuals nominated by the food safety authorities in the EU and EEA countries are granted access to the system. This rigorous access control process ensures a responsible chain of custody and security of the data stored within the system. Similar access control measures are implemented for the ECDC system in the public health domain, ensuring the confidentiality and integrity of the data across both systems.⁹

From a technical point of view, the EFSA One Health WGS system is a comprehensive platform operating on the EFSA cloud infrastructure. It consists of a user-friendly web application called the "WGS portal" and a Command Line Interface (CLI) application, both supported by a sophisticated, event-driven, backend infrastructure. This architecture enables users to interact with the system in their preferred manner, whether through the web portal or programmatically via the CLI. Moreover, the system offers dedicated Application Programming Interfaces (APIs) that facilitate seamless integration with existing systems utilized by European countries, promoting efficient communication and collaboration.⁷

Functionalities of the EFSA One Health WGS System and Its Interaction with the ECDC System

The EFSA One Health System stores core genome Multilocus Sequence Typing (cgMLST) profiles and various typing information derived from the genome sequences of different foodborne bacterial pathogenic isolates from food, feed, animals, and related environment. Users have the option to leverage EFSA's computing resources and bioinformatic pipeline to extract typing information. Alternatively, they can directly share their existing cgMLST profiles and typing data with the EFSA system, without the need to mobilize raw sequencing reads. The latter approach follows the principle of "move code to data," which means that EFSA publicly releases the bioinformatic pipeline for transparency and consistency. Users are free to choose whether to use EFSA's bioinformatic pipeline or their own internal pipeline, as long as they adhere to specific standards.⁷

In addition to the genomic information, users also have the opportunity to enrich the data by providing detailed contextual information about the samples from which the foodborne pathogens were isolated. These contextual data give valuable insights into the nature and origin of the samples, further enhancing the understanding of a foodborne outbreak.

The EFSA One Health WGS System and its counterpart at ECDC utilize dedicated protocols and APIs for machine-to-machine interaction, enabling automatic data exchange without human involvement. Genomic profile and contextual data are exchanged seamlessly when a verified correlation is found between reference human data and food and veterinary data (and vice versa), following established rules to ensure accuracy and relevance. This approach promotes smooth collaboration and data exchange across sectors. Moreover, users at the national level from multiple sectors can access joint cluster analysis, where relevant food and human data are visualized together, further enhancing insights and understanding. By combining genomic information across sectors with the respective detailed contextual data, the EFSA and ECDC systems enable a comprehensive analysis of foodborne outbreaks, thereby leading to better identification and tracking of potential risks to public health, enabling the implementation of mitigation strategies, and facilitating more effective response.¹⁰

In addition to their core functionalities (i.e., performance and visualization of joint cluster analysis), both the EFSA One Health WGS System and the ECDC system provide valuable components that enhance their capabilities. These systems are equipped with user-friendly and up-to-date business intelligence tools, enabling users to visualize the results of the cluster analyses and other general statistics, and ultimately to make better use of the data and strengthen their response to foodborne outbreaks. The visualizations generated by these tools provide intuitive representations of the analyses, enabling effective communication and collaboration among stakeholders and empowering users to gain deeper insights from the collected data. This, in turn, facilitates the understanding of complex patterns, leads to more informed risk assessments and improved response strategies for foodborne outbreaks, and supports informed decision-making.

Securing Trust: Confidentiality and Responsible Handling of Genomic Data

It is crucial to highlight that users maintain complete control over their data within the EFSA One Health WGS System. They have the authority to decide if and when they want to share their data with ECDC or other EFSA users. Additionally, users have the flexibility to edit or withdraw their data from the system at any time, maintaining full control over their data. This level of control ensures that users can manage their valuable information in accordance with their specific needs and requirements.^7,10

To further protect the confidentiality and integrity of the data stored within the system, strict access control rules and data protection mechanisms are in place. These rules ensure that users cannot see, modify, or download data that do not belong to their respective organization and sector. This safeguards against unauthorized access and use, and ensures that data remain secure within the systems. Furthermore, it is important to note that the data that are collected and stored in the system are to be used solely for the purpose of supporting multi-country foodborne outbreak detection and investigation and cannot be used outside this scope. By implementing these access control rules, data protection mechanisms, and usage restrictions, the EFSA One Health WGS System and the ECDC system ensure the confidentiality, integrity, and responsible handling of data, contributing to the overall trust and reliability of the system.¹⁰

Collaborative Efforts for Enhanced Foodborne Outbreak Detection and Assessment

Since its implementation, the EFSA One Health WGS System, along with its ECDC counterpart, has played a significant role in strengthening the European monitoring system for multi-country foodborne diseases. This system has facilitated collaboration among European countries and the two EU agencies, contributing to the formulation of hypotheses regarding the potential transmission of infections through food and to the assessment of multi-country clusters and outbreaks within the EU.

Serving as the central hub for sharing efforts among national food authorities, the EFSA system acts as the primary repository of genomic profiles in the food and veterinary sector at the European level. It plays a critical role as the single point for food and animal isolates' matching for ECDC, making it an essential component of the joint assessments conducted by ECDC and EFSA for foodborne illness outbreaks.¹¹

During this initial year of its implementation, the EFSA database has been regularly queried by ECDC to identify potential matches between food and human cases. This process has led to the discovery of multiple profiles of foodborne pathogens isolated from various food, animal, and environmental sources belonging to several clusters of human cases. Any significant findings resulting from these queries were evaluated, monitored, and thoroughly discussed in joint weekly meetings between EFSA and ECDC.

As a result of these collaborative efforts, EFSA and ECDC have published Rapid Outbreak Assessments and provided, when necessary, summaries to the European Commission and national-level risk managers regarding several events under monitoring.¹¹

The EFSA One Health WGS System: One Year in Numbers

One year since its release, the EFSA One Health WGS System database has accumulated approximately 1,300 genomic profiles of Listeria monocytogenes, Salmonella enterica, and Escherichia coli. These profiles were obtained from isolates collected from various sources including food, feed, animals, and the environment. Among the shared profiles, L. monocytogenes accounts for approximately 58 percent, followed by S. enterica at 40 percent, and E. coli at 2 percent.

In addition to the voluntary data shared by appointed users from EU and EEA countries, the EFSA database has been enriched with approximately 5,000 genomic profiles from isolates gathered in the food and veterinary domain. These profiles were retrieved from public repositories in a risk-based approach, with monthly updates. This import strategy implemented by EFSA aims to enhance the diversity of genomic profiles stored in the system, also enabling ECDC to conduct comprehensive searches that encompass both public and non-public data.

While the number of accumulated genomic profiles in the EFSA One Health WGS System database may be modest compared to the vast array of sequences available in public repositories, their value in responding to ongoing foodborne events and facilitating actions at the EU level remains exceptional. The majority of the 1,300 profiles were shared by 24 EU and EEA countries as part of their response to open investigations within the EU, showcasing a risk-based approach in addressing food safety concerns. This voluntary data sharing plays a critical role in enhancing the collective response to such events and promoting effective measures to safeguard public health.

However, it is noteworthy that over 50 percent of the data to the EFSA system database originates from the efforts of just three countries. This disparity highlights significant variations among EU and EEA countries in terms of their capacity and willingness to share data. It suggests that some countries are more active and engaged in data sharing, while others may face challenges or have different priorities that affect their contribution levels. Addressing these variations is crucial, and ongoing efforts are necessary to promote collaboration and encourage participation from all European countries.

Transitioning to a Proactive Data-Sharing Model

To unlock its full potential, the EFSA One Health WGS system should transition from a predominantly risk-based sharing model, which currently dominates data sharing in the EFSA system, to a proactive sharing approach across all European countries. Currently, only a few countries share data beyond specific foodborne illness events. By embracing this change, the EFSA and ECDC systems can significantly improve the efficiency and precision of identifying multi-country foodborne illness outbreaks before events are identified at the national level by proficiently clustering apparently unrelated human cases and isolates obtained from the food sector. However, this shift poses several challenges that need to be addressed.

Economic limitations, concerns over data ownership, and the need for clear guidelines on handling multiple joint clusters have been identified as major obstacles.¹² EFSA is actively collaborating with the network of EU experts to foster cooperation and promote equitable data-sharing practices. This includes establishing clear protocols, roles, and responsibilities. Furthermore, EFSA is committed to ensuring a more balanced contribution from all European countries and providing effective support and resources, when necessary.

By actively addressing these challenges, EFSA aims to advance toward a more efficient and proactive approach to data sharing. This will enable the system to make significant contributions to the protection of public health and the safety of food. It is imperative to overcome these obstacles and fully leverage the potential of the EFSA One Health WGS system for the benefit of all.

Note

The authors, Mirko Rossi, Eleonora Sarno, and Valentina Rizzi, are employed with the European Food Safety Authority (EFSA) in the BIOHAW Unit, which provides scientific and administrative support to the Panel on Biological Hazards in the area of hazard monitoring activities along the food chain from farm-to-fork. However, this article is published under the sole responsibility of the authors and may not be considered as an EFSA scientific output. The positions and opinions presented in this article are those of the authors alone and do not necessarily represent the views of EFSA or any official position or scientific works of EFSA. To learn more about the views and scientific outputs of EFSA, please consult the EFSA website at http://www.efsa.europa.eu.

References

EFSA Panel on Biological Hazards (EFSA BIOHAZ Panel). Koutsoumanis, K., A. Allende, A. Alvarez-Ordóñez, et al. "Whole genome sequencing and metagenomics for outbreak investigation, source attribution and risk assessment of food-borne microorganisms." EFSA Journal 3, no. 17 (December 2019): e05898. https://efsa.onlinelibrary.wiley.com/doi/full/10.2903/j.efsa.2019.5898.
Brown, B., M. Allard, M. C. Bazaco, J. Blankenship, and T. Minor. "An economic evaluation of the Whole Genome Sequencing source tracking program in the U.S." PLoS One 6, no 16 (October 2021): e0258262. https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0258262.
World Health Organization. "Whole genome sequencing for foodborne disease surveillance: landscape paper." Geneva, 2018. https://www.who.int/publications/i/item/789241513869.
Davis, C. T., J. Beal, L. Johnson, and T. Hadfield. "Genometrakr: Building a genomic epidemiology network for pathogen traceability." Foodborne Pathogens and Disease 17, no. 3 (2020): 141–144.
Nadon, C., I. Van Walle, P. Gerner-Smidt, et al. "PulseNet International: Vision for the implementation of whole genome sequencing (WGS) for global food-borne disease surveillance." Eurosurveillance 8, no. 23 (June 2017): 30544. https://www.eurosurveillance.org/content/10.2807/1560-7917.ES.2017.22.23.30544.
Shu, Y. and J. McCauley. "GISAID: Global initiative on sharing all influenza data—From vision to reality." Eurosurveillance 22, no. 13 (March 2017): 30494. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5388101/.
European Food Safety Authority (EFSA). Costa, G., G. Di Piazza, P. Koevoets, G. Iacono, E. Liebana, L. Pasinato, V. Rizzi, and M. Rossi. "Guidelines for reporting Whole Genome Sequencing-based typing data through the EFSA One Health WGS System." European Food Safety Authority (EFSA). EFSA Supporting Publication EN-7413. June 29, 2022. Mandate from European Commission. https://www.efsa.europa.eu/en/supporting/pub/en-7413.
Directive 2003/99/EC. Mandate M-2020-0015. "Art 31—Scientific and technical assistance." https://open.efsa.europa.eu/questions/EFSA-Q-2020-00101.
European Centre for Disease Prevention and Control (ECDC). "EpiPulse: The European Surveillance Portal for Infectious Diseases." June 22, 2021. https://www.ecdc.europa.eu/en/publications-data/epipulse-european-surveillance-portal-infectious-diseases.
EFSA, ECDC. "Collaboration agreement on the management and sharing of molecular typing data of isolates from human, food, feed, animal, and the related environment for public health purposes." 2022. https://www.efsa.europa.eu/sites/default/files/2022-06/collaboration-agreement-molecular-typing-EFSA-ECDC-WGS-DataCollection.pdf.
EFSA. "Food incident preparedness and response." April 17, 2023. https://www.efsa.europa.eu/en/topics/topic/food-incident-preparedness-and-response.
EFSA. "1st specific meeting on WGS, Scientific Network for Zoonoses Monitoring Data." February 27, 2023. https://www.efsa.europa.eu/en/events/1st-specific-meeting-wgs-scientific-network-zoonoses-monitoring-data.