Whole genome sequencing as a tool to strengthen foodborne disease surveillance and response: module 3: whole genome sequencing in foodborne disease routine surveillance: web annexes

Whole genome sequencing as a tool to strengthen foodborne disease surveillance and response Module 3. Whole genome sequencing in foodborne disease routine surveillance Web annexes

Whole genome sequencing as a tool to strengthen foodborne disease surveillance and response Module 3. Whole genome sequencing in foodborne disease routine surveillance Web annexes Whole genome sequencing as a tool to strengthen foodborne disease surveillance and response. Module 3. Whole genome sequencing in foodborne disease routine surveillance. Web annexes. ISBN 978-92-4-007242-8 (electronic version) © World Health Organization 2023 Some rights reserved. This work is available under the Creative Commons Attribution- NonCommercial-ShareAlike 3.0 IGO licence (CC BY-NC-SA 3.0 IGO; https:// creativecommons.org/licenses/by-nc-sa/3.0/igo). Under the terms of this licence, you may copy, redistribute and adapt the work for non- commercial purposes, provided the work is appropriately cited, as indicated below. In any use of this work, there should be no suggestion that WHO endorses any specific organization, products or services. The use of the WHO logo is not permitted. If you adapt the work, then you must license your work under the same or equivalent Creative Commons licence. If you create a translation of this work, you should add the following disclaimer along with the suggested citation: “This translation was not created by the World Health Organization (WHO). WHO is not responsible for the content or accuracy of this translation. The original English edition shall be the binding and authentic edition”. Any mediation relating to disputes arising under the licence shall be conducted in accordance with the mediation rules of the World Intellectual Property Organization (http://www.wipo.int/amc/en/mediation/rules/). Suggested citation. Web annexes. In: Whole genome sequencing as a tool to strengthen foodborne disease surveillance and response. Module 3. Whole genome sequencing in foodborne disease routine surveillance. Geneva: World Health Organization; 2023. Licence: CC BY-NC-SA 3.0 IGO. Cataloguing-in-Publication (CIP) data. CIP data are available at http://apps.who.int/iris. Sales, rights and licensing. To purchase WHO publications, see https://www.who.int/ publications/book-orders. To submit requests for commercial use and queries on rights and licensing, see https://www.who.int/copyright. Third-party materials. If you wish to reuse material from this work that is attributed to a third party, such as tables, figures or images, it is your responsibility to determine whether permission is needed for that reuse and to obtain permission from the copyright holder. The risk of claims resulting from infringement of any third-party-owned component in the work rests solely with the user. General disclaimers. The designations employed and the presentation of the material in this publication do not imply the expression of any opinion whatsoever on the part of WHO concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. Dotted and dashed lines on maps represent approximate border lines for which there may not yet be full agreement. The mention of specific companies or of certain manufacturers’ products does not imply that they are endorsed or recommended by WHO in preference to others of a similar nature that are not mentioned. Errors and omissions excepted, the names of proprietary products are distinguished by initial capital letters. All reasonable precautions have been taken by WHO to verify the information contained in this publication. However, the published material is being distributed without warranty of any kind, either expressed or implied. The responsibility for the interpretation and use of the material lies with the reader. In no event shall WHO be liable for damages arising from its use. This publication forms part of the document entitled Whole genome sequencing as a tool to strengthen foodborne disease surveillance and response. Module 3. Whole genome sequencing in foodborne disease routine surveillance. It is being made publicly available for transparency purposes and information. iii Contents Web Annex A. Working group terms of reference 1 Web Annex B. System description template 5 Web Annex C. Requirements document template 7 Web Annex D. Tool for mapping wet and dry lab sequencing capacities 9 Web Annex E. Outsourcing the wet lab component of WGS 11 Web Annex F. Establishing wet lab sequencing capabilities in the public health laboratory 17 Web Annex G. Outsourcing the bioinformatics component 27 Web Annex H. Establishing the bioinformatics component in the public health laboratory 35 Web Annex I. Epidemiologists and WGS during outbreak investigations 39 Web Annex J. The role of key stakeholders in WGS 45 Web Annex K. Business case template 49 Web Annex L. Template for cost estimates 51 Web Annex M. Key points to communicate to decision-makers 53 Web Annex N. Pilot study plan template 55 Web Annex O. Template for managing implementation 57 iv Whole genome sequencing as a tool to strengthen foodborne disease surveillance and response. Module 3. Whole genome sequencing in foodborne disease routine surveillance. Web annexes 1Web Annex A. Working group terms of reference When deciding whether to use whole genome sequencing (WGS) in the surveillance and response system for foodborne diseases, a working group of all of key stakeholders should be convened. To ensure the group’s productivity, clear terms of reference, roles and responsibilities should be spelled out for each member of the team (Table WA.1). A chairperson to facilitate meetings, and a secretary to record key decisions and actions from each meeting should also be appointed. 2Whole genome sequencing as a tool to strengthen foodborne disease surveillance and response. Module 3. Whole genome sequencing in foodborne disease routine surveillance. Web annexes Terms of reference The working group will bring together experts from relevant fields to provide inputs for the following steps. 1 Define the objectives of WGS within the outbreak response system. 2 Design a system that can incorporate WGS into the existing outbreak response, considering: wet and dry laboratory requirements A epidemiological requirementsB how WGS results will inform public health action.C 3 Write a description of how WGS will be incorporated into the outbreak response system. 4 Prepare a business case for senior policy-makers and officials to communicate the benefits and challenges of WGS in the proposed system. 5 Undertake a pilot project based on the description of the outbreak response system. 6 Evaluate the findings from the pilot project. 7 Make changes to the system’s design based on the pilot project’s results, and fully implement WGS. 3Web Annex A. Working group terms of reference Ta bl e W A .1 M em be r ( na m e) In st itu tio n Ro le Re sp on si bi lit ie s Ch ai r Fa ci lit at e m ee tin gs Se cr et ar y Ta ke m in ut es , in cl ud in g do cu m en tin g de ci sio ns a nd a ct io ns La bo ra to ry a dv ic e Pr ov id e ex pe rt a dv ic e on m at te rs re la te d to la bo ra to rie s Bi oi nf or m at ic s a dv ic e Pr ov id e ex pe rt a dv ic e on b io in fo rm at ic s an al ys es Ep id em io lo gi ca l a dv ic e Pr ov id e ex pe rt a dv ic e on h ow to in cl ud e ou tp ut s i n th e ou tb re ak in ve st ig at io n re sp on se sy st em Fo od sa fe ty a dv ic e Pr ov id e ex pe rt a dv ic e on m at te rs re la te d to fo od sa fe ty Te m pl at e: ro le s a nd re sp on si bi lit ie s o f w ho le g en om e se qu en ci ng w or ki ng g ro up m em be rs 4Whole genome sequencing as a tool to strengthen foodborne disease surveillance and response. Module 3. Whole genome sequencing in foodborne disease routine surveillance. Web annexes 5Web Annex B. System description template The working group will need to meet regularly and decide on some key aspects of how sequencing will fit within the surveillance and response system. The following template will provide some structure to help define the requirements within the system. Background What is WGS? What are the current surveillance conditions for the chosen pathogen(s)? Rationale Why is WGS important for this pathogen? What are the benefits of WGS for this pathogen? What are the short- and long-term goals of the country? Objectives of the surveillance and response system To monitor trends in the disease over time. To detect and respond to outbreaks of the disease. To evaluate the interventions to control the disease. [add any others that are relevant for your country] Data sources Describe the geographic area of coverage in your proposed surveillance system. Describe where specimens will come from (e.g. primary health care, hospitals, other). What are current follow-up public health practices regarding the collection of epidemiological data? For example, upon diagnosis of Listeria monocytogenes, a case interview is completed, regardless of typing; or for Salmonella spp., wait for clustering, and then interview cases in the cluster. 6Whole genome sequencing as a tool to strengthen foodborne disease surveillance and response. Module 3. Whole genome sequencing in foodborne disease routine surveillance. Web annexes Laboratory requirements Designated laboratory. Referral pathways, including sample transport. Number of samples anticipated. Technical laboratory requirements: - wet lab - dry lab - reagents, consumables - computing and information technology (IT) infrastructure - quality management - staff requirements - training staff. Reporting frequency and formats. This requirement should be decided in conjunction with epidemiologists in the public health authority. Data transfer and storage. Anticipated turnaround time from sample receipt. Public health requirements How will sequencing data be combined with epidemiological data? What are the thresholds for defining a cluster and taking public health action? Will the notifiable disease surveillance database need modification? Staffing requirements. Staff training. Links to public health action How will outbreaks be detected? Can data be shared with other sectors? If yes, indicate how data will be shared with the food safety and animal health sectors. What is the reporting frequency (combining sequencing and epidemiological information)? Managing transition in typing methods This section is only relevant to countries that are performing further typing of foodborne pathogens: describe how the transition will be managed. define the end date of traditional typing methods. 7Web Annex C. Requirements document template The requirements document will be provided to potential laboratories or other institutions being assessed when outsourcing sequencing and/or bioinformatic analyses during outbreak investigations. Most of the information will be outlined in the description of the outbreak response system, including a brief system overview (vision and objectives) and a brief description of what will be required, choosing: A to sequence samples only B to conduct bioinformatics analyses only, or C to conduct both sequencing and bioinformatics analyses. Detailed requirements 1 Sequencing A What is the sequencing frequency/capacity during an outbreak? B Is there enough capacity to increase the frequency or throughput if the outbreak becomes larger? 2 Bioinformatics A What is the assessment of sequencing quality? B What are the methods? C What are the outputs/reporting? D What is the turnaround time during outbreaks? E Are there additional analyses available on request (e.g. phylogenetic trees for clusters suspected to be an outbreak)? 3 Timeliness during outbreaks What is the timeframe for results (such as sample collection, processing, sequencing, data analysis) to be available to ensure rapid public health action? 4 Surge capacity and contingency plans What contingency plans are in place to ensure sequencing is not delayed? 5 Communication A Will results be required by a specific time each week? B Is there a process for prioritizing samples that need to be sequenced (e.g. during an outbreak)? C What are the communication channels for discussing WGS results between laboratory and public health staff? 8Whole genome sequencing as a tool to strengthen foodborne disease surveillance and response. Module 3. Whole genome sequencing in foodborne disease routine surveillance. Web annexes 9W eb A nn ex D . T oo l f or m ap pi ng w et a nd dr y la b se qu en ci ng c ap ac it ie s Ite m La bo ra to ry 1 (n am e) La bo ra to ry 2 (n am e) La bo ra to ry 3 (n am e) Li st th e pa th og en s t ha t c an b e se qu en ce d Sa lm on el la sp p. Sh ig a- to xi n pr od uc in g Es ch er ich ia co li Sh ig el la sp p. Li st th e w et la b pr oc es se s La bo ra to ry e qu ip m en t a nd st ep s n ec es sa ry fr om sa m pl e ar riv al to da ta p ro ce ss in g Se qu en ce r u se d Br an d an d m od el o f s eq ue nc er Co st o f w et la b pr oc es se s Co st p er sa m pl e Li st th e dr y la b pr oc es se s Q ua lit y c he ck s o f s eq ue nc in g da ta D at a pr oc es sin g fo r g en om e as se m bl y, w he n re qu ire d D at a pr oc es sin g fo r S N P id en tifi ca tio n, w he n re qu ire d Bi oi nf or m at ic s p ip el in es u se d N am e of so ftw ar e us ed Co st o f d ry la b pr oc es se s Co st p er sa m pl e Se qu en ci ng o ut pu ts Se qu en ce q ua lit y in fo rm at io n, M LS T, SN P tre e, o th er Ti m e be tw ee n re ce iv in g sp ec im en to se nd in g ou tp ut s to p ub lic h ea lth a ut ho rit ie s Ti m e in d ay s ( du rin g an o ut br ea k in ve st ig at io n) O ve ra ll c os t p er se qu en ce Co st p er sa m pl e [a dd o th er s s pe ci fic to la bo ra to ry , s ys te m , o r c ou nt ry us e] M LS T: M ul ti- lo cu s s eq ue nc e t yp e; SN P: si ng le n uc le ot id e p ol ym or ph ism . 10 Whole genome sequencing as a tool to strengthen foodborne disease surveillance and response. Module 3. Whole genome sequencing in foodborne disease routine surveillance. Web annexes 11 Web Annex E. Outsourcing the wet lab component of WGS In this annex, key considerations for outsourcing the sequencing component are discussed. There are several key steps, as shown in Fig. WE.1. Fig. WE.1. Steps in outsourcing the wet lab component of WGS Step 1 Develop a requirements document Step 2 Choose a laboratory to perform sequencing Step 3 Determine data ownership and data-sharing arrangements Step 4 Develop a contract for service 12 Whole genome sequencing as a tool to strengthen foodborne disease surveillance and response. Module 3. Whole genome sequencing in foodborne disease routine surveillance. Web annexes BOX WE.1 Considerations to include in a requirements document when outsourcing sequencing for routine surveillance Number of samples Estimate the number of samples that you will need sequenced per month. Estimate the required number of samples that would need to be sequenced during an outbreak (i.e. estimate the average size of outbreaks investigated in the past) or during seasonal peaks of disease. Specimen transport How will the samples get to the laboratory and in what form (culture, raw specimen, other)? How long will transport take? Are there companies that can transport samples? Will distance and temperature affect the samples? How will the samples be received at the WGS laboratory? Are there any legal or border security constraints related to sample transportation to the WGS laboratory (this may be important if the laboratory abroad)? Data attached to a sample Who owns the metadata attached to the sample (age, sex, postcode, result)? Timeliness To meet your objectives, what turnaround time do you need between sample delivery to the laboratory and availability of results? Develop a requirements document A requirements document sets out specific details of the service the outsourcing laboratory will provide. The more detail you can provide at this early stage will reduce frustrations and time delays during an actual outbreak situation. This document will also assist the potential outsourcing laboratories to provide accurate cost estimates and timeframes. The document should include estimates of the number of samples requiring sequencing per six months or yearly; the type of organisms likely to be sequenced; any possible seasonality to outbreak burden, etc. This information should be provided to the laboratory as a written form requirements document. Most of the information in the requirements document will come from the description of the system (Web Annex C). Box WE.1 contains a list of possible considerations. Step 1 13 Web Annex E. Outsourcing the wet lab component of WGS Choose a laboratory If you do not have existing collaborations with a sequencing laboratory or institute, make sure you have a valid and transparent laboratory selection process in place. It might help to set up a small committee to assist in identifying and assessing potential laboratories (Web Annex D). There are a number of ways in which sequencing can be outsourced, and a country may wish to investigate more than one option when determining suitability. A list of outsourcing options with their advantages and disadvantages is described in Table WE.1. Step 2 Table WE.1. Outsourcing option Advantages Disadvantages Inside the country Contract with university or other laboratory Builds national capacity Shorter distance for sample transportation The laboratory may have differing priorities and not understand the importance of timely results for public health Outside the country Contract with university or laboratory in another country Large sequencing facilities abroad might provide significantly lower prices due to wage differences and bulk reagent purchases Samples must be sent out of the country Increases transportation time and cost Potential customs issues Contract with a commercial laboratory Large sequencing facilities abroad might provide significantly lower prices due to wage differences and bulk reagent purchases Samples must be sent out of the country Increases transportation time and cost Potential customs issues Regional hub with multiple countries contributing specimens Allows comparison with contributing countries Samples must be sent out of the country Increases transportation time and cost Potential customs issues Advantages and disadvantages of outsourcing options for laboratory sequencing It is important to assess the laboratory capabilities and to be clear on what service and output they will provide. It is es- sential that the chosen laboratory has contingency plans in place for situations that would delay sequencing results such as power outages, staffing levels, access to reagents. The laboratory should have the capability to scale up in an outbreak situation. See Box WE.2 for sample questions to assess laboratory capacity. 14 Whole genome sequencing as a tool to strengthen foodborne disease surveillance and response. Module 3. Whole genome sequencing in foodborne disease routine surveillance. Web annexes Determine data ownership and data sharing arrangements Data ownership Establish data custodianship for sequences and what clearances are required if the outsourcing laboratory wishes to share the data on international databases or combine with their own data. Any reports/journal articles should be stipulated in the contract for service. Data sharing Guidance on data sharing is encouraged where possible (1) for human, animal, and food data. When outsourcing, it is important to seek legal advice concerning sample transport (especially if abroad), sharing information electronically with stakeholders and sharing information on international databases. Other international considerations When sending samples internationally, it will be important to adhere to International Air Transport Association Dangerous Goods Regulations (https://www.iata.org/en/programs/cargo/dgr/download/). The Nagoya Protocol on Access to Genetic Resources and the Fair and Equitable Sharing of Benefits Arising from their Utilization to the Convention on Biological Diversity is a supplementary agreement to the Convention on Biological Diversity (https://www.cbd.int/abs/about). However, it is unclear whether foodborne pathogens fall under this. Step 3 BOX WE.2 Questions for laboratory assessment How long will sequencing and the subsequent analysis take at the WGS laboratory? How will the laboratory prioritize outbreak samples for sequencing? Does the laboratory have contingency plans for equipment failures? Can the laboratory obtain the required reagents quickly enough to deal with a sudden increase in the number of samples to be sequenced? Are there any infrastructure issues that affect the laboratory that may have an impact on the timeliness of results (including power outages, staff on holidays, and other)? What are the capabilities to transfer raw sequence data? How does the laboratory control for cross-contamination? Is there a robust quality assurance programme in place with consistent documentation and results traceability? 15 Develop a contract for service Once a laboratory has been chosen, a service contract should be developed and signed by all parties. Box WE.3 provides some considerations to be included in the contract for service. The service contract should detail the requirements of the contractor and the service and outputs to be provided by the contractee. WHO has developed a material transfer agreement (MTA), available from: https://apps.who.int/blueprint/mta-tool/index.html. Step 4 Web Annex E. Outsourcing the wet lab component of WGS BOX WE.3 Contract for service when outsourcing the wet lab component of WGS Service contract inclusions Specification for the DNA sequencing quality. This includes how many ng of DNA in what volume to provide, if sending DNA rather than samples. Sequencing platform and setting that will be used. Specification of the quality of sequencing and assembly (e.g. minimum depth of 30X for phylogeny). File format for the output (e.g. fastq/fasta). How the files are delivered (e.g. via ftp/sftp). How long the link for accessing data will remain open (e.g. active link for one month, during which all data can be downloaded). Output size (e.g. 150 mega-basepair (Mbp) per isolate). Guaranteed turnaround time (e.g. 5 weeks). Cost. Develop an outsourcing requirements document. Enable public bidding on the contract, if required. Identify potential laboratories to conduct the sequencing. Assess and select a laboratory. Develop and sign a services contract. Update protocols with the information, including contact details, and any other pertinent matters. ACTION References 1. Guiding principles for pathogen genome data sharing. Geneva: World Health Organization; 2022 (https://apps.who.int/iris/ handle/10665/364222) accessed 19 April 2023). 16 Whole genome sequencing as a tool to strengthen foodborne disease surveillance and response. Module 3. Whole genome sequencing in foodborne disease routine surveillance. Web annexes 17 Steps in building in-country WGS capacity Step 1 Designate a laboratory Step 2 Plan the anticipated workflow Step 3 Choose a sequencing machine Step 4 Check the availability of reagents, consum- ables and equipment Step 5 Carry out quality assurance Fig. WF.1. Web Annex F. Establishing wet lab sequencing capabilities in the public health laboratory This annex documents all of the key considerations for building capacity in the public health laboratory to perform the wet lab component of WGS. There are several key steps, as shown in Fig. WF.1. A country does not need to complete one step before moving onto another step. Often, some of the steps will run in parallel. For example, when choosing a sequencing machine it will be important to understand the reagents that are required and the infrastructure. This may then influence the decision about which laboratory will be designated with the responsibility to perform WGS. The main point is that all of the steps are important when making decisions about who will conduct WGS and how. 18 Whole genome sequencing as a tool to strengthen foodborne disease surveillance and response. Module 3. Whole genome sequencing in foodborne disease routine surveillance. Web annexes Designate a laboratory For a country wishing to establish wet lab WGS capabilities, there will be some important considerations when designating a public health laboratory to perform WGS. Uninterrupted power supply. If this is not possible, consider options such as installation of uninterrupted power supply units. Appropriate temperature and humidity in laboratory environments, typically in the 20–25 °C range and 30– 70% relative humidity range. Stable internet connection with adequate bandwidth for processing and sharing sequences. Is in a well-connected location, easy to reach from the rest of the country and from outside the country, i.e. connected to main roads and, ideally, close to international airports. This measure will enable improved access to: - maintenance and repair services offered by equipment - suppliers of reagents and consumables - technical support and troubleshooting - sample transport (if applicable). If the coverage of the surveillance system is the whole country, it may be necessary to have multiple laboratories performing sequencing. If there are multiple laboratories who will contribute, it will be important to harmonize testing methods and reporting to ensure the results are comparable. You will need to organize a workshop to bring all the laboratories together to decide on testing methods. Referral pathways must be documented. Ensure that the diagnostic laboratories are aware of where to send the samples. Ensure a well-organized sample forwarding service if outsourcing is foreseen as part of the system, especially if service providers are outside of the country. In this respect, restrictions to the international exchange of pathogens, cross-border bureaucratic issues and difficulties with temperature- controlled shipments must be properly addressed. Step 1 Designate a laboratory in-country to perform the WGS.ACTION 19 Web Annex F. Establishing wet lab sequencing capabilities in the public health laboratory Plan the anticipated workflow For the chosen pathogen and area of surveillance system coverage, examine the notifiable disease surveillance system database to determine the number of cases per year that have been notified, over the years the data are available. If possible, calculate the 5-year average, including the range. From the average, it is then possible to estimate the number of samples for sequencing expected each week (it may be necessary to account for seasonality). Ensure the number of outbreaks of the pathogen(s) have been reported and investigated and where possible, know the size of the outbreaks to plan for surge capacity in outbreak situations. It will also be necessary to revisit the objectives for surveillance to determine the required frequency of sequencing runs. If one of the objectives is to identify outbreaks, turnaround time should be quick to enable investigations and the rapid implementation of control measures. Turnaround time in the laboratory will be determined by the number of samples that can be sequenced in one run, and by the pathogen’s incubation period. Pathogens with longer incubation periods (e.g. Listeria monocytogenes or hepatitis A virus), may not require weekly sequencing, but the ability to scale up in an outbreak situation must exist. A pathogen such as Salmonella, which is outbreak-prone, will require a more rapid turnaround time to be able to detect outbreaks early. From these considerations, it will be possible to estimate the number of samples to be sequenced over a defined period, as required in the surveillance system. This will be useful when planning sequencing workflows with the necessary throughput in the following section. Step 2 Estimate the number of samples (e.g. on a weekly basis) and the sequencing frequency required according to the surveillance objectives. Define the optimal WGS throughput combining number of samples and required frequency of sequencing runs. ACTION Choose a sequencing machine When choosing a sequencing machine, matching the needs of the system to the equipment is important to avoid wasting resources. Some key considerations when choosing a sequencer are: - sample throughput (average and maximum per run) - sequencing run time (i.e. the time it takes for a sequencing machine to complete a sequencing run) - instrumentation and service costs - reagent costs - infrastructure needs - skills and experience of staff. Step 3 20 Whole genome sequencing as a tool to strengthen foodborne disease surveillance and response. Module 3. Whole genome sequencing in foodborne disease routine surveillance. Web annexes Choose a sequencer to match your needs.ACTION At present, sequencing machines are very diverse with respect to throughput. There is generally a minimum number of isolates that can be sequenced per run (depending on the pathogen), for that run to be economically feasible, unless substantial extra-costs per isolate can be accepted. Running WGS with high frequency (e.g. once or twice a week) is key to achieve timely genomic information in a timely manner. Considering that timeliness is a feature that strongly impacts surveillance performance, choosing a sequencer that allows optimal frequency (with optimal throughput) is paramount for high-performance of WGS in the surveillance system. A discussion on the different types of sequencers available and the technical specifications are given in the WHO landscape paper on whole genome sequencing for foodborne disease surveillance (1). Sequencers with different throughputs do exist on the market, and the choice of a sequencing machine with the right throughput for a given surveillance setting is possible, but other features can complicate the choice. Generally, sequencers with low throughput have considerably higher sequencing costs per genome compared with high throughput machines. High throughput machines however in general have longer run times which may not be feasible for foodborne disease surveillance. In addition, they would need a high number of samples to be sequenced simultaneously to be economically feasible. This can create budgetary problems that negatively impact surveillance programs and must be carefully taken into account. Moreover, different machines based on different technologies could create issues of standardization and cross-validation of results. The extra costs of validation and standardization have to be considered as a managerial factor while setting up WGS within the surveillance system. In addition to correct identification of machine throughput, laboratories performing WGS for surveillance will need to keep up with changing technology. The rapidly evolving nature of WGS technology requires careful evaluation of investments with the awareness that purchased equipment could become obsolete quickly and that the evolution of equipment and methodologies could change the laboratory setting and workflow in the short-to-medium term. At the same time, despite the evolving nature of WGS technology, standardization and quality assurance of surveillance activities based on WGS need to be guaranteed. To get the most from the investment in WGS equipment, a country can: choose equipment based on defined throughputs, while taking into account the cost per genome, and balance between cost per genome and appropriate frequency of sequencing runs; and consider the rapid evolution of WGS technology and manage the laboratory assuming short-to-medium lifecycle of sequencers and hardware. 21 Web Annex F. Establishing wet lab sequencing capabilities in the public health laboratory Availability of reagents, consumables and equipment WGS of foodborne pathogens requires various specific reagents, consumables and pieces of ancillary laboratory equipment, such as centrifuges, incubators, fluorometers, thermocyclers, and autoclaves. Purchasing the necessary reagents and equipment can be a challenge in countries where there is limited or no commercial coverage by companies. Furthermore, some of the reagents have limited shelf-life, preventing laboratories from maintaining large stocks over long periods of time. Sequencing centres need to purchase reagents frequently and close to the place of usage in order to deliver timely results and minimize the waste of expensive materials due to expiry. Moreover, the reagents need to be stored at refrigeration temperatures. National/international bureaucracy can also be a barrier in the purchasing process. Altogether, these conditions can make the timely and economical availability of reagents and hardware difficult. Countries and/or institutions in charge of WGS for foodborne outbreak investigations can use the following strategies to maximize reagent, consumable and equipment use. Organize the workflow and throughput of sequencing and data analysis to be as constant as possible. This means defining the reagent and equipment use and planning ahead for the number of samples to be run each week and the frequency of sequencing runs. It is then possible to plan for when reagents and consumables need to be re-ordered. This will guarantee timely processing of specimens and results for use in the surveillance system. If access to supplies is an issue, consider reaching out to international partners, neighbouring countries or other sequencing centres in-country to look at bulk purchasing of reagents and equipment (Case study A). It may also be possible to partner locally with laboratories with existing capacity, to help reduce costs (Case study B), or to explore the use of alternate or in-house prepared reagents. To optimize reagent use and safeguard the quality of results, the reagents need to be stored correctly. In this respect, reagent handling and storage procedures should be documented in a laboratory manual and strictly observed. The same is necessary for the maintenance of sequencing and ancillary laboratory hardware. The sequencer and ancillary equipment will also need to be regularly serviced to prolong the life of the equipment, ensure it works when it needs to and at the required performance level. Step 4 22 Whole genome sequencing as a tool to strengthen foodborne disease surveillance and response. Module 3. Whole genome sequencing in foodborne disease routine surveillance. Web annexes Case study A Apply economies of scale: network-based negotiating and purchasing power One strategy to navigate expensive equipment and reagents and supply monopolies is to combine the purchasing of multiple institutions. This potentially lowers the cost of supplies (due to the higher volumes purchased), and enables a stronger voice in negotiations. Example: World Health Organization support for negotiations Brief description of how WHO was able to support improvements to tuberculosis diagnostics and enable their implementation in many countries by negotiating concessional pricing for the diagnostic supplies (2). Case study B Finding local partners to reduce costs in Colombia If enough resources are not available to your programme, seek partnerships with outside collaborators who are well funded and whose goals complement yours. Colombia’s example Colombia established an integrated food chain surveillance system to monitor antimicrobial resistance (AMR) among key pathogens, including Salmonella and Campylobacter strains. It was possible to bring together all relevant partners in the public and private sector and determine where samples would be collected from, how they were tested, reported and analysed. A pilot programme was run for 4 years and was able to demonstrate the scale of AMR. The building of partnerships was vital to obtaining funding for the pilot study (3). 23 Web Annex F. Establishing wet lab sequencing capabilities in the public health laboratory When supply issues are difficult to overcome due to irregular throughput of specimens in WGS centres or difficulties in guaranteeing reagent supply, some potential solutions include reducing the number of laboratories performing WGS within a single country or a group of countries to ensure there is one laboratory with adequate throughput. This will require: identification of the single laboratory that will perform the testing; clear specimen referral pathways, including any bureaucratic requirements for specimen transport outside of the country; and collaborative agreements between countries about when and how the isolates will be sequenced and the information that will be shared. Plan anticipated workflow and throughput to estimate weekly equipment and reagents requirements. Develop an ordering schedule based on how long it takes for reagents and consumables to arrive at the laboratory from the time of ordering. This may include bulk purchasing with neighbouring countries to minimize cost. Document reagent handling and storage in the laboratory procedures manual to minimize wastage. In the laboratory procedures manual, document the maintenance and servicing frequency required for the sequencer and ancillary laboratory equipment. ACTION 24 Whole genome sequencing as a tool to strengthen foodborne disease surveillance and response. Module 3. Whole genome sequencing in foodborne disease routine surveillance. Web annexes Quality assurance When WGS is used as a subtyping method for foodborne infections, its accuracy and precision should be assured to guarantee the reliability of the results reported to the surveillance system. A robust quality assurance programme is important for any diagnostic and public health laboratory. Relevant international standards encompassing all steps of WGS for surveillance of infections are currently being worked on. Therefore, laboratory methods and procedures developed to date are not internationally standardized and validated. Similarly, the processes of data analysis and storage have not been standardized, which hampers complete comparability of results across countries. Furthermore, validation schemes of analysis pipelines for WGS data are difficult to develop, due to the diversity of WGS platforms used to generate data and the various analytical strategies in use. To overcome the challenges posed by the lack of international standards for quality assurance in WGS, a country can undertake the following actions. Adopt proper quality assurance for laboratories performing WGS and associated data processing. The adopted framework of quality assurance should comply with the relevant international standards (International Organization for Standardization (ISO)/International Electrotechnical Commission (IEC) 17025 for testing laboratories). In addition to implementation of quality assurance, accreditation of laboratories by the national accreditation body (if it exists) is advised. Whenever specific standards for WGS analysis have not been issued by national or international bodies, develop and validate the required standards in the form of internal methods and procedures, according to the quality assurance system implemented. As an alternative and/or complement to the development of internal standards, a country can join existing WGS surveillance networks that have already developed or are in the process of developing the necessary standards. The standards can be shared among the network partners and can provide access to technical expertise and best practice models. Review information about the global microbial identifier proficiency test 2017 at https://www. globalmicrobialidentifier.org/workgroups/about-the-gmi-proficiency-test-2017. Some of the international bodies who have developed standards include: the European Union research initiative, COMPARE PulseNet International GenomeTrakr. Once a quality assurance system has been established and internal standards have been developed, a country can take part in proficiency testing, which includes both WGS and related data processing through bioinformatics pipelines. This provides an opportunity to check and align internal standards and quality assurance. Joining relevant networks can facilitate this step. Validate any internally developed standards and/or participate in the validation of externally developed standards. This can be done best by participating in proficiency testing. Participate in training activities organized by countries and institutions that have already developed expertise in WGS surveillance. Although this does not directly compensate for the lack of standards, training of staff is a critical component of quality management and can contribute to the overall improvement of quality assurance. This is especially the case with technologies, like WGS, that significantly depend on skilful staff for reliable results. Joining relevant networks can facilitate this step. Step 5 25 Web Annex F. Establishing wet lab sequencing capabilities in the public health laboratory Request ISO or other relevant standardization bodies to undertake the development of proper standards and/or guidance documents in the field of laboratory quality management, analytical methods, data analysis and validation processes with regard to WGS. The steps a country can take to ensure accurate and reliable sequencing results are summarized in Fig. WF.2. Fig. WF.2. Steps to ensure quality assurance in the laboratory Step 1 Adopt quality assurance that complies with international standards (ISO/IEC 17025) Step 2 Create internal standards for WGS methods and document in the laboratory manual Step 3 Join WGS surveillance networks Step 4 Participate in proficiency testing to validate internal standards Step 5 Ensure regular training of staff performing sequencing and generating reports ISO/IEC 17025 Step 6 Request ISO or other international stan- dardization body to undertake develop- ment of proper internal standards 26 Whole genome sequencing as a tool to strengthen foodborne disease surveillance and response. Module 3. Whole genome sequencing in foodborne disease routine surveillance. Web annexes Access to troubleshooting to support WGS To use WGS for surveillance purposes, it is important that there is access to troubleshooting. This is particularly critical with sequencing, where issues can occur with equipment, laboratory methods and reagents. Furthermore, the specificity of some of the reagents and hardware requires specialized support staff with significant technical expertise. Certain countries or areas could lack customer service/support from reagent and equipment suppliers or, more generally, they could have difficulties in accessing assistance and technical support from companies and peers. If a country is unable to access expertise and knowledge when issues are experienced with sequencing, there is a loss of timeliness, effectiveness and reliability in the analysis process. Countries or their institutions planning to undertake surveillance using WGS should be aware of the barriers represented by difficulties in accessing troubleshooting locally. To overcome this limitation, the following strategies may be used. Carefully evaluate the actual availability of support. If the current support is inadequate, proper measures should be taken to overcome this barrier. Local troubleshooting conditions could be negotiated with suppliers through purchase agreements. For example, a clause could be negotiated to specify that ongoing troubleshooting support will be provided. Coordination with other countries within the region that share the same problems could strengthen the above negotiations or create a common reference support centre at the regional level. Collaboration with countries or institutions with significant experience in WGS could be undertaken, for instance, by twinning projects for the implementation and development of WGS for surveillance. Such partnerships can be particularly helpful, especially in the early stages of WGS implementation. References 1. Whole genome sequencing for foodborne disease surveillance: landscape paper. Geneva: World Health Organization; 2018 (https://apps.who.int/ iris/handle/10665/272430, accessed 11 April 2022). 2. Xpert MTB/RIF assay for the diagnosis of pulmonary and extrapulmonary TB in adults and children: policy update. Geneva: World Health Organization; 2013 (https://apps.who.int/iris/handle/10665/112472, accessed 11 April 2022). 3. Donado-Godoy P, Castellanos R, León M, Buitrago G, Tafur MA, Byrne BA et al. The establishment of the Colombian integrated program for antimicrobial resistance surveillance (COIPARS): a pilot project on poultry farms, slaughterhouses and retail market. Zoonoses Public Health. 2015;62(Suppl. 1)58–69. Bibliography • Expert Opinion on the introduction of next-generation typing methods for food- and waterborne diseases in the EU and EEA. Solna: European Centre for Disease Prevention and Control; 2015 (https://www.ecdc.europa.eu/en/publications-data/expert-opinion-introduction-next- generation-typing-methods-food-and-waterborne, accessed 12 April 2022). • Next generation sequencing implementation guide. Silver Spring: Association of Public Health Laboratories; 2016. 27 Web Annex G. Outsourcing the bioinformatics component This annex documents all of the key considerations for outsourcing the bioinformatics component of WGS, which is solely a dry lab component. There are several key steps, as shown in Fig. WG.1. Fig. WG.1. Steps in outsourcing bioinformatics component of WGS Step 1 Develop a requirements document Step 2 Choose an institution to perform bioinformatics analysis Step 3 Determine data ownership and data-sharing arrangements Step 4 Develop a contract for service 28 Whole genome sequencing as a tool to strengthen foodborne disease surveillance and response. Module 3. Whole genome sequencing in foodborne disease routine surveillance. Web annexes Develop a requirements document A requirements document specifies in detail the service the outsourcing party will provide. The more detail you can provide at this early stage will reduce frustrations and time delays during an actual outbreak situation. This document will also assist the potential outsourcing laboratories to provide accurate cost estimates and timeframes. When contacting potential laboratories (or other institutions), an estimate of specific needs will need to be provided, in order to get accurate costing and timeframes. This will include: an estimate of the number of samples requiring sequencing every six months or yearly; the type of organisms likely to be sequenced, any possible outbreak burden seasonality, etc.; the required outputs for outbreak response purposes (e.g. phylogenetic trees with single nucleotide polymorphism (SNP) analysis); and expected median and maximum turnaround times for results (during an outbreak situation). This information should be provided to the laboratory in a written form requirements document. Web Annex C contains a template for the requirements document. Box WG.1 contains a list of of possible considerations for the requirements document. Step 1 29 Web Annex G. Outsourcing the bioinformatics component Technical considerations for conducting the bioinformatics component Number of samples Estimate the number of samples that may need to be analysed during an outbreak (based on the minimum and maximum from last two years of outbreaks). Estimate how many samples will need to be analysed in a year. Raw sequence data transfer How will the raw sequence data be transferred and who is responsible for it? Are there any legal constraints from transmitting data to the potential party (this may be important if the party is situated outside the country)? Outputs What is the output from the bioinformatics analyses that will be provided to public health authorities? How comparable is this format with the current public health database? Data attached to sample sequences What metadata will be attached to the sequences? Who owns the metadata attached to the sample (age, sex, postcode, result)? If there is limited metadata attached to the WGS outputs, how will it be linked to epidemiological data? Timeliness To meet your objectives, what turnaround time do you need from sample delivery to the laboratory to when the results are received by public health authorities? BOX WG.1 30 Whole genome sequencing as a tool to strengthen foodborne disease surveillance and response. Module 3. Whole genome sequencing in foodborne disease routine surveillance. Web annexes Choose an institution to perform bioinformatics analyses If you do not have existing collaborations with a bioinformatics institute, when choosing an appropriate institution make sure you have a valid and transparent process in place. It would be beneficial to form a small committee to help identify and assess potential institutions. Web Annex D contains a template that can be used to assess potential institutions. There are several ways in which bioinformatics analyses can be outsourced, each with advantages and disadvantages (Table WG.1). A country may wish to investigate more than one option when determining suitability. Step 2 Table WG.1 Outsourcing option Advantages Disadvantages Inside the country Contract with a university or other laboratory Builds national capacity Slowly helps build capacities in bioinformatics in the public health laboratory The institution performing bioinformatics analyses may have differing priorities and not understand the importance of timely results for public health Outside the country Contract with a university or institution abroad Can access bioinformatics experts in situations where capacity may not exist Data-sharing concerns with the outsourced institution running the analyses Contract with a commercial company Can access bioinformatics experts in situations where capacity may not exist Established protocols and experience Data-sharing concerns with the outsourced institution running the analyses Regional hub with multiple countries contributing specimens or isolates Allows comparison with contributing countries Data-sharing concerns with the outsourced institution running the analyses Advantages and disadvantages of outsourcing options for bioinformatics analyses It is important to assess the institute’s capabilities and to be clear on what service and output they would provide to you. It is essential that the chosen institute has contingency plans in place for situations that would delay analysis such as power outages, staffing levels, and access to reagents. Box WG.2 lists potential questions that a country can assess the institute against when outsourcing the bioinformatics component. 31 Web Annex G. Outsourcing the bioinformatics component Questions for Bioinformatics Institution Assessment Bioinformatics How is sequence contamination detected using bioinformatics? What tools are used and how? Does the institution have staff or is the work outsourced? What analysis procedures do they use? Are the analysis methods validated? Are the analysis methods compatible with your colleagues that you share and compare data with? How will the data be provided back to you? In what timeframe will the results be provided to the public health authorities? Can the results be provided in a format that is compatible with the surveillance database and/or a format that epidemiologists can use to inform public health action? Is there capacity to deal with ad-hoc requests (e.g. during outbreaks, request for different analyses, confirmation of results that inconsistent with the epidemiological data) Who owns the sequencing data? How will the data be stored? Will the sequence data with patient information be stored securely? Interpretation of results Will this be a joint process between bioinformaticians and public health staff? If so, how will it be conducted? What epidemiological information is required to help refine the bioinformatics analysis? What epidemiological data will be shared with the laboratory and in what format? BOX WG.2 32 Whole genome sequencing as a tool to strengthen foodborne disease surveillance and response. Module 3. Whole genome sequencing in foodborne disease routine surveillance. Web annexes Determine data ownership and data sharing arrangements Data ownership For WGS to be fully utilized the genomic data needs to be combined with epidemiological data. With previous typing methods the laboratory could provide a number such as pulsed field gel electrophoresis (PFGE) or a serogroup with the sample ID; however, WGS requires epidemiological data to determine appropriate cluster thresholds. This should be a collaborative process to decide where thresholds are set, which may require sharing data. Establish data custodianship for sequences and what clearances are required if the outsourcing laboratory wishes to share the data on international databases or combine with their own data. Any reports/journal articles should be stipulated in the contract for service. Data sharing Currently, guidance on sharing results and information is limited for human data, animal data and food data. When outsourcing it important to seek legal advice concerning the sharing of information electronically with stakeholders, and sharing information on international databases. Table WG.2 provides examples of metadata that may be found on international databases along with raw sequence data. Step 3 Table WG.2 Variable Description Sample ID number Unique sample identifier Strain ID Alpha numerical, laboratory identifier e.g. MLS-4810-04 Genus, species, subspecies, serovar A description of the microbe that has undergone sequencing (e.g. Salmonella enterica subsp enterica Montevideo) Collected by Name of the organization/institution/laboratory submitting the sequence to the database Sample collection date Dates of sample collection (DD/MM/YYYY) Country Country where the sample originated Sample type This is a description about the source of the specimen/sample, such as human, food, environment, animal, or water Isolate contributor Contact details for the person submitting the data from the organization mentioned in ‘Collected by’ field (this is generally not published, but is used as contact information in case a matching strain is observed elsewhere in the world) Examples of metadata that may be uploaded to international databases along with raw sequence data 33 Develop a contract for service Once an institution has been chosen, a service contract should be developed and signed by all parties. The service contract should detail what is required of the contractor, as well as the service and outputs to be provided by the contractee. Box WG.3 includes important specifications to include in a contract for service of the dry lab component of WGS. Step 4 BOX WG.3 Contract for service when outsourcing the bioinformatics component of WGS Service contract inclusions Bioinformatics analysis to be performed. This may include serotype, core genome multilocus sequence type (cgMLST)/whole genome multi-locus sequence type (wgMLST), AMR genes, aligned DNA to the reference genome. Format of the analysis outputs. How the analysis outputs will be delivered. Estimated turnaround time for bioinformatics analysis (e.g. 1 week). Cost. Develop an outsourcing requirements document. Identify potential institutions for outsourcing. Assess and select an institution. Develop and sign a services contract. ACTION Web Annex G. Outsourcing the bioinformatics component 34 Whole genome sequencing as a tool to strengthen foodborne disease surveillance and response. Module 3. Whole genome sequencing in foodborne disease routine surveillance. Web annexes 35 Web Annex H. Establishing the bioinformatics component in the public health laboratory Bioinformatics is a highly specialized area which is required for any laboratory wishing to use WGS for routine surveillance in the long- term. There will need to be significant investments in either hiring a bioinformatician within the public health laboratory, or in training laboratory staff to be able to interpret the outputs from bioinformatics analyses correctly. This is a long-term process which will take time. Further reading on bioinformatics in public health are provided at the end of this web annex. Two key areas that need to be developed for the dry lab component of WGS for outbreak investigations are: computer hardware, networking and data storageA bioinformatics analyses.B Computer hardware, networking and data storage WGS generates large amounts of data that need to be processed and stored, requiring high-performance computing capability and data storage facilities. Also necessary is a reliable, uninterrupted, high-speed internet connection, with sufficient bandwidth to transfer and analyse large amounts of data. Planning for storage capacity should be based on the anticipated number of weekly sequences generated. Once a sequence is generated, it will need to be stored for future analysis and comparison. As sequencing improves and expands, it will be necessary to increase the storage capacity. It is important to anticipate when extra data storage will be required to prevent running out of space, as not to delay the laboratory’s ability to sequence. Options for data storage include: purchasing storage space in a locally hosted server with a well developed intranet to move data around; purchasing storage space from cloud providers; collaboration agreements with internal or external partners with data storage space, for example, with a regional computing centre; and using data storage in publicly available databases, such as the National Centre for Biotechnology Information, DNA Databank of Japan and the European Nucleotide Archive. Computing capacity also needs to be considered, but it is generally only an issue if sequencing more than 100 isolates per week (1) or if a large number of samples are included in the analysis. Collaborative agreements with internal or external partners (e.g. universities or research institutes inside or outside of the country) may be useful to access computing resources. 36 Whole genome sequencing as a tool to strengthen foodborne disease surveillance and response. Module 3. Whole genome sequencing in foodborne disease routine surveillance. Web annexes Research internet providers in the country to ensure a reliable, high-speed internet connection that has the appropriate bandwidth. Estimate data storage requirements for 1 and 5 years. Make sure that adequate data storage is available, whether it needs to be purchased or can be accessed through collaborative agreements. ACTION Bioinformatic analyses Bioinformatics pipelines are automated computational workflows that combine multiple programmes running in tandem, written specifically to analyse the sequences and provide outputs that can be used by public health authorities to determine how closely related isolates are. Some pipelines can also report on the presence or absence of AMR and/or virulence genes and mutations. To generate the required outputs from sequencing, the read information needs to be (1): 1 analysed for quality, making sure that sequencing generated information can be further analysed 2 assembled to reconstruct the genome 3 aligned to a reference genome for a comparative analysis, such as SNP analysis 4 analysed to derive subtyping results, for example, a core genome or whole genome MLST 5 used to build phylogenetic trees or transmission trees using WGS data. To be able to generate the outputs from sequencing, a country needs to have: 1 access to bioinformatics tools, pipelines and software (the WHO landscape paper on whole genome sequencing for foodborne disease surveillance (1) provides details about the bioinformatics tools currently available); 2 personnel to perform the analyses and guide the interpretation; and 3 personnel to maintain and upgrade tools and applications. Given that bioinformatics is a highly specialized field, it may be difficult to recruit bioinformaticians. It may be possible to recruit mathematics, computer science or statistics graduates with training in computational methods applied to other sectors (such as banking or finance). Options for countries without easy access to bioinformaticians include: reaching out to international partners who can provide bioinformatics support purchasing licenses for commercial products in collaboration with internal or external partners establishing regional centres where a bioinformatician can service multiple countries in the region. 37 Web Annex H. Establishing the bioinformatics component in the public health laboratory Choosing an appropriate reference genome If one of the outputs will be an SNP analysis, ensuring access to appropriate reference genomes to enable comparative analysis is crucial. This type of analysis is considered highly discriminatory, capable of identifying the smallest differences between samples. As such, SNP-based approaches are useful in the investigation of foodborne outbreaks where deciphering the exact correlations between cases and sources of infection is critically important in order to assign cases to the respective outbreaks and to precisely attribute outbreaks to their sources. The identification of SNPs within a group of samples that are moderately to highly related to each other, as is the case in an outbreak, is performed by comparing each genome to a high-quality reference genome, i.e. one that is well curated; ideally closed; and as similar as possible to the genomes under investigation. A good reference genome can be obtained by generating it from a sample known to be highly related to the samples under investigation (e.g. one of the outbreak samples) or retrieving it from a publicly available repository. Generating a reference genome requires equipment, reagents and expertise that are generally beyond the standard of basic WGS laboratories. To overcome this barrier, collaboration with other institutions that have the necessary equipment and expertise might be established, or the task could be outsourced. Additional constraints are the lack of WGS laboratories in the country or area and the high cost of generating high-quality reference genomes, both in terms of consumables and time, especially when high-quality draft genomes have to be assembled to use as reference. The alternative is to retrieve already existing reference genomes from accessible repositories, which is generally exempt of cost and infrastructure constraints. Nevertheless, it might be difficult to find the reference needed and there could be uncertainty regarding its closeness to the samples under investigation. Determine how to access bioinformatics support, by hiring a bioinformatician, training laboratory staff to interpret bioinformatics analyses, or participating in international networks for bioinformatics advice. Determine which analytics programmes to use. Ensure sufficient computing capacity and internet connectivity. ACTION References 1. Whole genome sequencing for foodborne disease surveillance: landscape paper. Geneva: World Health Organization; 2018 (https://apps.who.int/iris/handle/10665/272430, accessed 11 April 2022). Bibliography • Expert Opinion on the introduction of next-generation typing methods for food- and waterborne diseases in the EU and EEA. Solna: European Centre for Disease Prevention and Control; 2015 (https://www.ecdc.europa. eu/en/publications-data/expert-opinion-introduction-next-generation-typing-methods-food-and- waterborne, accessed 12 April 2022). • Next generation sequencing implementation guide. Silver Spring: Association of Public Health Laboratories; 2016. • Oakeson KF, Wagner JM, Mendenhall M, Rohrwasser A, Atkinson-Dunn R. Bioinformatic analyses of whole- genome sequence data in a public health laboratory. Emerg Infect Dis. 2017;23(9):1441–45. doi.org/10.3201/ eid2309.170416. 38 Whole genome sequencing as a tool to strengthen foodborne disease surveillance and response. Module 3. Whole genome sequencing in foodborne disease routine surveillance. Web annexes 39 Web Annex I. Case studies This annex contains case studies from some of the countries who implemented WGS for the routine surveillance of foodborne pathogens. This information was current at the time of developing this Web Annex. US Centers for Disease Control and Prevention PulseNet USA has integrated MLST, including strain allelic profile nomenclature into its surveillance database that is linked to databases/systems with epidemiological information including the System for Enteric Disease Response, Investigation and Coordination (SEDRIC) (1) and the Listeria initiative, and for resistance monitoring, the National Antimicrobial Resistance Monitoring System for enteric bacteria (2). WGS has replaced PFGE as the preferred tool in PulseNet for outbreak detection and investigation. The primary outputs from WGS are phylogenetic trees based on cg and wgMLST and allele codes for strain nomenclature. Information about serotype, virulence profile, antimicrobial resistance and plasmid profiles is also entered into the PulseNet database from WGS. Allele codes are automatically transferred into outbreak line lists. A case-case comparison tool assessing exposures from the Listeria monocytogenes initiative to different nodes on a phylogenetic tree is used in investigations of listeriosis outbreaks. PulseNet includes WGS is an integral part of the USA foodborne surveillance systems nationally as well as locally. It is used in the detection and investigation of both multistate and local outbreaks by the United States Centers for Disease Control and Prevention (US CDC) and local public health and food regulatory laboratory end-users. Food regulatory agency laboratories in the USA including the United States Food and Drug Administration (FDA) and the United States Department of Agriculture (USDA) Food Safety and Inspection Service (FSIS) are participants in PulseNet, and use the information generated in the network routinely for regulatory purposes. WGS implementation has not changed this routine which has been in place for more than 20 years with PFGE. 40 Whole genome sequencing as a tool to strengthen foodborne disease surveillance and response. Module 3. Whole genome sequencing in foodborne disease routine surveillance. Web annexes Public Health England Public Health England (PHE) uses WGS for routine surveillance and detection of outbreaks caused by foodborne pathogens. Depending on the specific pathogen, the WGS is used for the following outputs: 1 K-mer based species identification ID (https://github.com/phe-bioinformatics/kmerid); 2 MLST (https://github.com/phe-bioinformatics/MOST); 3 in silico Serotype (SeqSero / SSI database); 4 virulence profile (e.g. stx subtype) (https://github.com/flashton2003/stx_subtyping) and E. coli pathogroup markers (e.g. aggR, eae, ipaH, LT, ST); 5 SNP address (https://github.com/phe-bioinformatics/snapperdb); and 6 AMR. All the WGS data are linked to an in-house database called the Gastro Database Warehouse (GDW). This allows the WGS data to be linked to a patient ID, along with relevant epidemiological data (e.g. age, sex, postcode, travel history or clinical data if included on lab submission form). For outbreak detection, SNP cluster reports are generated for all the samples using SNP profiles. Clusters are updated weekly, and the output discussed jointly with laboratory, epidemiology, and bioinformatics personnel. There is a standard operating procedure for the generation of SNP cluster reports along with example outputs. In addition, phylogenetic trees are also provided to aid in PHE’s investigation. In summary, WGS for public health surveillance was implemented to facilitate outbreak detection and investigation, and to provide comprehensive data on highly discriminatory typing, virulence factors, and AMR, in order to monitor trends and emerging threats to public health. WGS outputs are linked to patient data and basic epidemiological information via the GDW. Each week the data are analysed and summarized in a spreadsheet, and appropriate short-term public health action in response to outbreaks and incidents is discussed by laboratory, epidemiology, and bioinformatics personnel. Data informing regulatory decision making are analysed as part of the function of the national reference laboratory and national epidemiology team together, on a monthly or quarterly basis, and feedback to users and policy decision-makers in the form of reports, presentations and publications. 41 Web Annex I. Case studies Public Health Agency of Canada WGS outputs used for cluster detection of foodborne diseases The primary WGS tool currently being used for cluster detection for Salmonella spp. and Listeria monocytogenes strains is wgMLST. The National Microbiology Laboratory (NML) may also choose to run SNP analyses to confirm results when dealing with a new serotype, or at the beginning of a national outbreak investigation. Identified clusters are posted with a tree/phylogeny, as well as a line list, which will typically contain a cluster code, and in the case of Listeria monocytogenes, the wgMLST nomenclature assigned through BioNumerics. The use of WGS routinely for isolates of E. coli and Shigella was anticipated for late 2018. Where appropriate, an assessment of the similarity to known travel clusters, existing clusters being monitored or actively investigated, or closed clusters that may have a known/suspected exposure is also provided. WGS results from the laboratory The NML identifies WGS clusters, assigns cluster codes and posts clusters on the Canadian Laboratory Surveillance Network (CLSN). CLSN is a secure platform housed in the Canadian Network for Public Health Intelligence; it is accessible to laboratories and epidemiologists. The Outbreak Management Division (OMD) receives a notification each time an update is posted to CLSN. WGS results are not automatically integrated with OMD event database. Instead, the event database was modified to accommodate WGS outputs (such as case counts, isolation dates, allele ranges). The NML also provides the Foodborne Disease and AMR Surveillance Division, on a weekly basis, a line list of the sequenced isolates, whether from human, food, animal or environmental sources, which includes dates of isolation, metadata and cluster codes, if any. This database is updated on a weekly basis, and will form the basis for conducting surveillance monitoring activities as WGS implementation moves forward. WGS outbreak detection WGS analysis is performed weekly at the NML, and discussed jointly between NML staff and epidemiologists. The laboratory process: A is an overall visual assessment of a wgMLST tree containing all existing sequence data to identify groupings of isolates with a high degree of similarity; B generates focused trees for each individual grouping/cluster of samples; and C assesses allele differences – typically interested in isolates falling within 0–10 alleles of one another. May expand/ contract the range depending on the circumstances, as each situation is different. Laboratory criteria for cluster detection For Listeria monocytogenes: two or more isolates that are within 0–10 alleles within 120 days. For Salmonella spp.: two or more isolates in 60 days that are within 0–10 alleles for serotypes other than Enteritidis, Heidelberg and Typhimurium. For Enteritidis, Heidelberg, and Typhimurium, three or more isolates that are within 0–10 alleles in 60 days. 42 Whole genome sequencing as a tool to strengthen foodborne disease surveillance and response. Module 3. Whole genome sequencing in foodborne disease routine surveillance. Web annexes Development of criteria OMD use the criteria developed by NML. NML criteria is based on a combination of retrospective and prospective data. A retrospective WGS study of a number of known clusters/outbreaks with an epidemiological link to a confirmed source, for the top five enteric pathogens was completed, as well as studies for specific, and common, PFGE pattern combinations, as well as data collected from concurrently sequencing isolates through the course of an ongoing outbreak investigation. Data were reviewed by both laboratory and epidemiologists. All data showed that confirmed cases in outbreaks with a definitive source were typically within 0–10 alleles/SNPs. This range of interest is under continual assessment as we collect prospective WGS data, and may be revised as we move forward, but has thus far been found to be a reasonable guideline for true cluster detection. Prioritization of clusters for follow-up Epidemiologists from the OMD review all new and updated clusters as they are posted, and meets with NML and surveillance partners each week to review and discuss clusters of interest (e.g. new clusters, clusters that require follow- up and updates on clusters previously followed up). OMD epidemiologists developed WGS assessment criteria/ considerations that are used to assess and prioritize clusters. Considerations include serotype and frequency, number of isolates included in the cluster, allele range and branching, demographic profiles, presence of non-clinical isolates (e.g. food samples), and isolation dates, among others. The OMD assessment process and considerations have developed and changed as the understanding of WGS has increased. Evaluation of the role of WGS in the surveillance system Although a formal evaluation has not been conducted at the time this report was prepared, the impact of WGS on Listeria monocytogenes in Canada was reviewed by analysing the number of clusters identified in previous years using PFGE alone with those identified once transitioned to WGS. Depending on the country and the levels of human illness related to the different organisms, the number of clusters detected will vary. For example, while the number of listeriosis clusters in the United States of America increased once WGS was implemented, in Canada, this was not the case, with a far lower number of clusters identified in 2017, when WGS was implemented, as compared with previous years. The OMD has begun work on defining cluster metrics, for use in future evaluation activities. Epidemiologists needed for the transition to WGS It was not necessary to hire more epidemiologists with the transition to WGS. The internal operational model was revised to create a consistent team that is dedicated to assessing clusters and detecting potential outbreaks. The role of the OMD Database Manager became very important with the transition to WGS, in terms of supporting modifications to the event database and the development of specific outputs (e.g. a weekly WGS cluster summary). Approach to upskilling existing epidemiology workforce for WGS NML developed and delivered training materials and sessions on WGS methods, interpretation and impacts. This training was developed collaboratively with the epidemiologists in the surveillance and outbreak groups to ensure that the information was not too technical and applied to the work conducted by epidemiologists. One of the training sessions provided by NML focused on questions submitted by epidemiologists, to ensure that the training addressed issues that were emerging through the transition and working with WGS information. The OMD and the Foodborne Disease and AMR Surveillance Division have also attended various seminars offered by other organizations (e.g. Cornell University, US CDC). Much of the training has occurred on the job, and through weekly meetings with NML colleagues, who provide valuable insights on the interpretation of laboratory data. 43 Web Annex I. Case studies Malbran Institute At the time of reporting, Malbran Institute in Argentina was in the process of building a national WGS database to monitor Salmonella, Shigella, E. coli, and Vibrio cholerae, and WGS was not used in routine surveillance. However, WGS has been used in targeted investigations of Salmonella, Shigella and E. coli outbreaks. In that context, WGS data are linked to epidemiological information obtained from the National Health Surveillance System (SNVS, by its Spanish acronym). WGS outputs are integrated to data from Surveillance Systems at the National Reference Laboratory, and the resulting data goes to local epidemiology agencies, and to the National Epidemiology Department of the Ministry of Health for further decision making. In outbreak investigations, a reference-based SNP tree is created. AMR genes or mutations in different databases and virulence genes are also considered when epidemiologically relevant. Foodborne disease surveillance data are included in the SNVS, through the National Laboratories Network, and the Mandatory Infectious Diseases Reporting System, mainly for monitoring and surveillance, and to support outbreak detection and investigation. The SNVS was in the process of moving into a new system “SISA”, for a better integration of the individual information with other agencies (animal, agriculture and food). Outbreak investigations start at the local level with epidemiology, food safety and human laboratory sectors (primary isolation, identification and characterization) using PFGE and WGS (when possible). These functions are supported by the National Reference Laboratory and National Epidemiology Department, when required. All data are incorporated by each user into the SNVS to centralize the information. WGS is in the initial steps of incorporation into the National System. The Malbran Institute can provide support or troubleshooting advice for WGS in a national surveillance network framework.. As a reference laboratory for PulseNet Latin America and the Caribbean, the Institute’s mission is to provide support to the region in subtyping techniques. Furthermore, as an external laboratory of the GenomeTrakr Network the Institute has been using SNP trees to monitor Salmonella, Shigella sonnei, and E. coli. to monitor clonal distribution. 44 Whole genome sequencing as a tool to strengthen foodborne disease surveillance and response. Module 3. Whole genome sequencing in foodborne disease routine surveillance. Web annexes PulseNet international At the time this case study was written, the PulseNet International Network had grown to encompass 86 countries in seven regions. Its goal is to build capacity for a global foodborne surveillance and outbreak detection system. For PulseNet International to be successful in WGS implementation using a gene-by-gene analysis approach (wgMLST), the network is developing standardized laboratory and analytical methods, a strain nomenclature system, and a platform for sharing data generated for public health activities. To be able to compare different bacterial isolates and their subsequent whole genome sequence data across borders, a set of predefined quality metrics, assembly and allele calling algorithms, cg/wgMLST allele databases, and strain nomenclature must be agreed upon. These standards will be agreed upon globally and will be validated to meet the needs of foodborne disease surveillance and outbreak detection. PulseNet International agreed to one scheme for each organism and identified hosts for the databases and tools to satisfy the current data submission policies related to what data the participatory countries can legally share in public domain. While all surveillance networks may decide which platform and methodologies work best for them, PulseNet is committed to using a centralized wgMLST database with limited metadata for its mostly patient-derived isolates. References 1. Standardized Epidemiologic and Laboratory Protocol for Investigation of Outbreaks: Sedric [website]. Atlanta: Centers for Disease Control and Prevention; 2022 (https://www.cdc.gov/foodsafety/outbreaks/investigating-outbreaks/sedric.html, accessed 31 May 2023). 1. National Antimicrobial Resistance Monitoring System (NARMS) [website]. Atlanta: Centers for Disease Control and Prevention; 2023 (https://www.cdc.gov/narms/index.html, accessed 31 May 2023). 45 Web Annex J. The role of key stakeholders in WGS This section has been adapted from the WHO landscape paper (1). It describes the roles of the three key stakeholders in using WGS for the surveillance and response system for foodborne diseases. The main three roles required for WGS to be used effectively in the surveillance and response system are the molecular microbiologist, bioinformatician, and epidemiologist. While each role is detailed below, many tasks overlap, and should be conducted jointly by these professionals. 46 Whole genome sequencing as a tool to strengthen foodborne disease surveillance and response. Module 3. Whole genome sequencing in foodborne disease routine surveillance. Web annexes Molecular microbiologist The molecular microbiologist is responsible for initiating and conducting sequencing, which may include the following tasks. Initial phenotypic and molecular identification and characterizations of isolates including culture purification and storage. Genomic DNA extraction purification, and library preparation with appropriate quality controls. Sequencing run set-up on the sequencing machine. Downloading sequencing data and reviewing quality measurements for the run. Maintaining accurate secure records of all procedures, including electronic databases of genome sequences and related data. Appropriate record-keeping and accounting, and maintaining all equipment and consumables. Maintaining culture collections to facilitate retrospective audits and selection of internal control strains for WGS experiments. Participating in quality control and assurance programmes to ensure national and international harmonization of sequencing methods. In collaboration with the bioinformatician, identifying reference genomes when required and determining the requirements for an IT environment that supports genomic data sharing, storage and archiving. In collaboration with the epidemiologist, determining what data will be required for validating interpretation criteria and for performing cluster assessments. This might include studies where isolates from well defined outbreaks (i.e. epidemiological evidence is strong and/or a source was identified) are sequenced retrospectively to help the validation process. In collaboration with the epidemiologist, identifying potential challenges in implementation, processes to be used, and who/how/when will the isolates be sequenced to better interpret surveillance information. 47 Web Annex J. The role of key stakeholders in WGS Bioinformatician The bioinformatician is responsible for post sequencing data analysis, and genomic data storage. Routine data analyses can sometimes be performed by the molecular microbiologist, which may include the following tasks. Managing the computational analysis of sequencing data. This is usually in collaboration with molecular microbiologists who are domain experts in genomics of particular pathogens. Available options range from off-the-shelf software solutions, and online tools to in-house or open source pipelines. Implementing, verifying and managing computer-based algorithms for genome assembly, variant detection and isolate clustering through the use of phylogenetic tree construction. Maintaining accurate secure records of all procedures, including electronic databases of genome sequences and related quality control data. Assessing the quality of original and processed sequencing data. In collaboration with the molecular microbiologist and epidemiologist, determining the most appropriate method of analysis (e.g. de novo assembly, mapping to a reference, etc.). In collaboration with the epidemiologist and molecular microbiologist, developing reports and cluster- naming conventions. In collaboration with the molecular microbiologist and epidemiologist, assessing clusters to determine which ones will be followed up and turned into outbreak investigations. Supporting epidemiologists and public health staff in understanding bioinformatic analyses outputs. 48 Whole genome sequencing as a tool to strengthen foodborne disease surveillance and response. Module 3. Whole genome sequencing in foodborne disease routine surveillance. Web annexes Epidemiologist The epidemiologist is responsible for collecting epidemiological information and integrating it with WGS data, which may include the following tasks. In collaboration with the molecular microbiologist and bioinformatician, determining cluster nomenclature and the most appropriate reporting method. Defining what constitutes a cluster (e.g. four Salmonella notifications in the previous two weeks with related genomic sequences) to support epidemiological investigations. Determining how WGS data will be implemented in the existing public health surveillance infrastructure, and evaluate the implementation, in order to identify challenges, needs and gaps. Determining reporting formats for WGS outputs for routine surveillance and outbreak investigations, e.g. line lists only, line lists and trees, trees only, other. Identifying database needs for incorporating WGS outputs into existing surveillance systems. Combining epidemiological information with reported WGS data. Identifying cases needing follow-up to collect epidemiological information, including what samples are part of the cluster. An outbreak may consist of more than one distinct genomic sequence. Ensuring that appropriate information is collected to fulfil legal and legislative requirements. In conjunction with the molecular microbiologist and bioinformaticians, developing and evaluating data- sharing protocols that meet legal and legislative requirements. Monitoring WGS timelines to ensure appropriate and timely public health action. Reference 1. Whole genome sequencing for foodborne disease surveillance: landscape paper. Geneva: World Health Organization; 2018 (https:// apps.who.int/iris/handle/10665/272430, accessed 11 April 2022). 49 Web Annex K. Business case template This annex is a template that might be helpful for developing the business case. To get started, convene a working group to analyse the template and complete the applicable sections. All service plans should state the objectives of sequencing regarding the surveillance and response system; describe how the samples and data will be analysed and used in surveillance and response; describe the resources required, and an estimated budget, which may also include a cost–benefit analysis. Background Using plain language, explain sequencing. Rationale for strengthening WGS within the surveillance and response system This section should include a discussion of the benefits of WGS. Ideally, use local examples or examples from similar countries in your region. If neither are available, an international example could be useful to demonstrate the benefits of WGS. Current status of WGS This should be a broad statement about the current national capacity for WGS related to outbreak detection and response to foodborne diseases. Describe how the current system meets all the minimum requirements for the selected option, then detail current activities in relation to WGS and how they contribute to surveillance and outbreak responses. Approach to strengthening WGS within the surveillance and response system Describe the chosen implementation option. It can also include the vision from the relevant option. You can cut and paste this over from the guidance document and then modify it to meet the vision within the country. Some preliminary work with stakeholders may be required to ensure a common vision for WGS in the country. Stakeholders Identify key stakeholders in using WGS in the surveillance and response system. These will be individuals in the working group who developed the description of the surveillance system, such as laboratory staff, bioinformaticians that will support the analyses, the public health authorities who will receive the data (i.e. the epidemiologists). Stakeholders may also include relevant experts from the food safety and animal health sectors, to ensure the establishment of integrated food chain surveillance applying a One Health approach. 50 Whole genome sequencing as a tool to strengthen foodborne disease surveillance and response. Module 3. Whole genome sequencing in foodborne disease routine surveillance. Web annexes Specific requirements for WGS in the surveillance and response system This section will describe the higher-level actions needed during implementation to help the country towards achieving its visions and meeting the objectives of the surveillance and response system, including: the pathogens under surveillance and why they were chosen the geographic region under surveillance an estimate of throughput where sequencing will be conducted personnel requirements training requirements. Transition from traditional typing methods Only use this section in countries that are conducting further typing of foodborne pathogens. Include: how transition will be managed the length of time traditional typing methods and WGS will run in parallel. Budget estimate for WGS for the surveillance and response system Include a table with the costs of using WGS in the surveillance and response system. Include a cost–benefit analysis, if it was part of the planning process. Timelines for implementation Document key milestones anticipated during implementation and when they are likely to be completed. Risks Document any potential risks that may stall or hamper implementation of WGS. Sustainability plan Indicate other priority areas that would benefit from the implementation of WGS once it is established for foodborne diseases. If establishing capacity within the public health system, point to any opportunities to recuperate funds by charging research agencies for WGS services. Indicate what current laboratory tests might be replaced by WGS if used routinely. Determine whether there is a regular budget line for funding WGS as the method for subtyping foodborne diseases. Evaluation Indicate that there will be a pilot study first, which will be evaluated. Include a statement about how the country intends to evaluate WGS within the surveillance and response system. Document the expected time for the evaluation, and whether there are any conditions linked to the outcomes from the evaluation. 51 W eb A nn ex L . T em pl at e fo r c os t e st im at es Th is te m pl at e h as b ee n de sig ne d fo r c ou nt rie s t o us e w he n th ey ar e d ev elo pi ng a bu sin es s c as e a nd w an t t o pr ov id e a co st es tim at e f or th e e sta bl ish m en t o f W G S a nd o ng oi ng m ain te na nc e c os ts. Th e i ni tia l c os t c ov er s p ur ch as es an d th e o ng oi ng co st re fer s t o th e p ric e o f m ain te na nc e o r o ng oi ng p ro fes sio na l d ev elo pm en t s uc h as u pd at in g s ta ff tra in in g. Ite m In iti al C os t O ng oi ng C os ts (p er y ea r) O ut so ur ci ng Se qu en ci ng (p er is ol at e) Bi oi nf or m at ic s La bo ra to ry e qu ip m en t Se qu en ce r An ci lla ry e qu ip m en t La bo ra to ry re ag en ts a nd co ns um ab le s Re ag en ts G la ss w ar e, p la st ic w ar e, o th er m at er ia ls Co m pu te rs a nd n et w or ki ng Co m pu te rs Re lia bl e in te rn et co nn ec tio n D at a st or ag e Bi oi nf or m at ic s An al yt ic to ol s, pi pe lin es , e tc . H um an re so ur ce s St affi ng (n um be r o f n ew st aff to b e re cr ui te d) Tr ai ni ng e xi st in g st aff Ad di tio na l c os ts TO TA L Te m pl at e 52 Whole genome sequencing as a tool to strengthen foodborne disease surveillance and response. Module 3. Whole genome sequencing in foodborne disease routine surveillance. Web annexes 53 Web Annex M. Key points to communicate to decision-makers The following talking points can be adapted to communicate with decision-makers regarding WGS and how it can be used to strengthen foodborne disease surveillance and response. High resolution Public health professionals can use WGS data to quantify differences (and similarities) between individual strains of a microbe more accurately than other tools available to date. This means clusters detected are more likely to represent cases exposed to the same source, outbreaks are more likely to be solved and resources can be focused where they are likely to have the most effect. Additionally, there are greater strength of evidence when a food isolate matches human cases. Earlier detection of outbreaks With greater discrimination to determine relatedness, putative outbreaks can be detected early, when a small number of cases with pathogens sharing the same or nearly identical sequence can be observed. The experience of some countries using WGS for surveillance purposes is that more foodborne outbreaks are detected, with fewer cases in each outbreak (1). Sequence data can be used for multiple purposes Once a sample has been sequenced, that sequence can be stored electronically and analysed in different ways without additional laboratory work, including: typing/cluster identification AMR testing virulence gene detection comparison with samples from other countries (e.g. GenomeTrakr) characterization of new strains/novel pathogens research to better understand the epidemiology or microbiology of the pathogen. Cost–benefits superior to traditional typing methods In 2016, the economic return on investment using PFGE is approximately US$ 70 for every US$ 1 invested by public health agencies. This demonstrates a significant economic and public health benefit from the system, and these impacts are expected to be even greater with the superior performance of WGS (2). Comparability over time Over time, WGS bioinformatic analyses and outputs may change, but the actual sequence will remain available for future use, offering opportunities to understand the epidemiology of foodborne diseases in greater detail, both currently and into the future. 54 Whole genome sequencing as a tool to strengthen foodborne disease surveillance and response. Module 3. Whole genome sequencing in foodborne disease routine surveillance. Web annexes Improves information collected as part of outbreak investigation WGS can also enable robust case ascertainment during an outbreak investigation (3). Sequencing information can show isolates are genetically closely related when epidemiological information may be lacking or where exposure information cannot be ascertained through interview. It can provide clues for public health staff to reinterview the cases or ascertain exposures in a different ways. Enhanced food safety system When combined with data from epidemiological investigations, WGS can provide strong evidence that a particular food or environmental source is linked to foodborne diseases in humans. Information gathered from WGS for surveillance and outbreak investigations can be a driver to allocate further resources to strengthen the food control system (4). Support other disease priority areas Once implemented for foodborne pathogens, WGS can be used to support other disease groups. The sequencing machine is the same for all pathogens. Bioinformatics will be different, depending on the analysis pipeline used but the process is the same. Additional advantages to using WGS Provides greater understanding of the epidemiology of pathogens under surveillance. Informs policy-making on antimicrobial utilization in a clinical setting for improved patient care. Informs policy-making on antimicrobial utilization in animals used for human food production, when data from across the food chain are integrated (4). References 1. Allard MW, Strain E, Melka D, Bunning K, Musser SM, Brown EW et al. Practical value of food pathogen traceability through building a whole-genome sequencing network and database. J Clin Microbiol. 2016;54(8):1975–83. doi.org/10.1128/JCM.00081-16. 2. Scharff RL, Besser J, Sharp DJ, Jones TF, Peter GS, Hedberg CW. An economic evaluation of PulseNet: a network for foodborne disease surveillance. Am J Prev Med. 2016;50(5 Suppl. 1):S66–S73. doi.org/10.1016/j.amepre.2015.09.018. 3. Butcher H, Elson R, Chattaway MA, Featherstone CA, Willis C, Jorgensen F et al. Whole genome sequencing improved case ascertainment in an outbreak of Shiga toxin-producing Escherichia coli O157 associated with raw drinking milk. Epidemiol Infect. 2016;144(13):2812–23. doi.org/10.1017/ S0950268816000509. 4. Tang KL, Caffrey NP, Nóbrega DB, Cork SC, Ronksley PE, Barkema HW et al. Restricting the use of antibiotics in food-producing animals and its associations with antibiotic resistance in food-producing animals and human beings: a systematic review and meta-analysis. Lancet Planet Health. 2017;1(8):e316–e327. doi.org/10.1016/S2542-5196(17)30141-9. 55 Web Annex N. Pilot study plan template The following template may be useful in the implementation of the pilot study for a trial of WGS during a foodborne diseases outbreak investigation. In completing the template, you will document how the pilot study will function; decide what questions need to be answered; set key milestones and deliverables; and indicate how results will be communicated. It will also define the criteria used to evaluate the pilot study and to translate the findings into full-scale implementation, consistent with the description of the outbreak response system. Governance Describe who participates in the advisory group (more senior decision-makers) and who is part of the pilot study team (technical staff for implementation). Pilot study objectives Document the objectives of the pilot study. These will be different from those of surveillance and response system. Typically, the pilot objectives focus on evaluating the approach in preparation for full- scale implementation. It would also be beneficial to state the pilot study questions in more detail. Pilot study design Define the scope of the pilot study, making decisions based on the surveillance system description (i.e. if the pilot will take place in a specific area or whether a single pathogen will be chosen). Explain the pilot study method, for example, whether it will be retrospective or prospective. It may be helpful to include a flowchart of how the samples will flow through the system during the pilot study. Milestones and deliverables List anticipated milestones for the pilot study. For example: • pilot study plan drafted • sequencing of samples begins • end date for sequencing • evaluation report with recommendations for future implementation • list deliverables, including dates. 56 Whole genome sequencing as a tool to strengthen foodborne disease surveillance and response. Module 3. Whole genome sequencing in foodborne disease routine surveillance. Web annexes Communication Document key communication exchanges that will take place during the pilot study. This should include: • the frequency of the advisory group and pilot study team meetings • the frequency of results reported from the laboratory • who will receive the reports. Evaluation criteria Document the criteria that will be used to assess the success of the pilot study. Describe how the evaluation will take place. 57 Web Annex O. Template for managing implementation Once the working group has been established and key decisions are made, it will be possible to start recording actions and developing the implementation plan. This annex contains a template to keep track of all of the tasks required when designing and implementing WGS for foodborne disease surveillance purposes. As a country decides whether to outsource or develop capacity within the public health laboratory for the wet lab and dry lab components, actions from each of the relevant annexes will need to be added to this template to build the implementation plan. The steps for completing the implementation plan within the working group are as follows. Take the template as a starting point. Step 1 Review the template, deleting or adding tasks to the list. Step 4 Once you have decided whether to outsource or develop capacity in the public health laboratory for the wet lab component of WGS, copy the actions from the relevant annex and add them to this template. For out- sourcing the wet lab, use the actions in Web Annex E. For establishing wet lab capacity in the public health laboratory, use the actions in Web Annex F. Step 2 Once you have decided whether to outsource or develop capacity in the public health laboratory for the dry lab component of WGS, copy the actions from the relevant annex and add these actions to the template. For outsourcing the dry lab, use the actions in Web Annex G. For establishing dry lab capacity in the public health laboratory, use the actions in Web Annex H. Step 3 58 Whole genome sequencing as a tool to strengthen foodborne disease surveillance and response. Module 3. Whole genome sequencing in foodborne disease routine surveillance. Web annexes Ac tiv iti es Ta sk s Pe rs on re sp on si bl e Ye ar Ye ar Ja n Fe b M ar Ap r M ay Ju n Ju l Au g Se p O ct N ov D ec Ja n Fe b M ar Ap r M ay Ju n Ju l Au g Se p O ct N ov D ec Co nv en e a w or ki ng g ro up o n im pl em en tin g W G S fo r o ut br ea k in ve st ig at io n of a nd re sp on se to fo od bo rn e di se as es Id en tif y st ak eh ol de rs Es ta bl ish a w or ki ng g ro up D oc um en t t er m s o f r ef er en ce D efi ne ro le s a nd re sp on sib ilit ie s D ec id e w he th er W G S is a pp ro pr ia te Re vi ew th e m in im um re qu ire m en ts fo r W G S in th e In tro du ct or y m od ul e As se ss th e ca pa ci tie s i n th e de ci sio n tre e (F ig . 3 o f t he In tro du ct or y m od ul e) W or ki ng g ro up to c on si de r w hi ch im pl em en ta tio n op tio n is m os t ap pr op ri at e Re vi ew th e op tio ns fo r i m pl em en tin g W G S in th e In tro du ct or y m od ul e U se th e de ci sio n tre e in th e In tro du ct or y m od ul e to a ss ist in ch oo sin g an o pt io n Te m pl at e 59 Web Annex O. Template for managing implementation Ac tiv iti es Ta sk s Pe rs on re sp on si bl e Ye ar Ye ar Ja n Fe b M ar Ap r M ay Ju n Ju l Au g Se p O ct N ov D ec Ja n Fe b M ar Ap r M ay Ju n Ju l Au g Se p O ct N ov D ec D ev el op a d es cr ip tio n of h ow W G S w ill b e in co rp or at ed in a fo od bo rn e ou tb re ak in ve st ig at io n En su re y ou h av e th e ap pr op ria te te ch - ni ca l s ta ff to d ev el op th e ou tb re ak in ve st ig at io n sy st em d es cr ip tio n W or k th ro ug h th e ou tb re ak in ve st ig at io n sy st em d es cr ip tio n te m pl at e (W eb A nn ex B o f t hi s m od ul e) D efi ni ng g oa ls a nd o bj ec tiv es D efi ne th e sh or t- an d lo ng -te rm se qu en ci ng g oa ls D ec id e w hi ch p at ho ge n w ill be se qu en ce d D efi ne th e ob je ct iv es o f u sin g W G S to en ha nc e ou tb re ak in ve st ig at io ns o f fo od bo rn e di se as es D ec id e th e ar ea o f c ov er ag e fo r t he su rv ei lla nc e sy st em W et la b D ec id e w he th er to o ut so ur ce w et la b or b ui ld ca pa ci ty in p ub lic h ea lth la bo ra to ry [if o ut so ur ci ng w et la b, in se rt a ct io ns fr om W eb A nn ex E h er e] [if b ui ld in g ca pa ci ty fo r w et la b in th e pu bl ic he al th la bo ra to ry , in se rt a ct io ns fr om W eb An ne x F he re ] D et er m in e an d do cu m en t s am pl e re fe rra l p at hw ay s En su re sa m pl es a re tr an sp or te d co rre ct ly 60 Whole genome sequencing as a tool to strengthen foodborne disease surveillance and response. Module 3. Whole genome sequencing in foodborne disease routine surveillance. Web annexes Ac tiv iti es Ta sk s Pe rs on re sp on si bl e Ye ar Ye ar Ja n Fe b M ar Ap r M ay Ju n Ju l Au g Se p O ct N ov D ec Ja n Fe b M ar Ap r M ay Ju n Ju l Au g Se p O ct N ov D ec D ry la b D ec id e w he th er to o ut so ur ce d ry la b or b ui ld c ap ac ity in p ub lic h ea lth la bo ra to ry [if o ut so ur ci ng d ry la b, in se rt a ct io ns fr om W eb A nn ex G h er e] [if b ui ld in g ca pa ci ty fo r d ry la b in th e pu bl ic he al th la bo ra to ry , in se rt a ct io ns fr om W eb An ne x H h er e] D et er m in e an d do cu m en t w he re bi oi nf or m at ic a na ly se s w ill be co nd uc te d Re po rt in g to th e ou tb re ak in ve st ig at io n sy st em La bo ra to ry a nd p ub lic h ea lth st aff de ci de : • ho w to re po rt re su lts fr om W G S to p ub lic h ea lth a ut ho rit ie s • an a gr ee ab le fr eq ue nc y fo r re po rt in g W G S re su lts to p ub lic he al th a ut ho rit ie s • ho w to in te rp re t t he re su lts a nd re po rt tr en ds o ve r t im e Pu bl ic h ea lth a ut ho rit ie s m od ify th e su rv ei lla nc e da ta ba se to ca pt ur e ag re ed W G S ou tp ut s En su re th e da ta d ic tio na ry fo r t he su rv ei lla nc e da ta ba se is u pd at ed to re fle ct th e ch an ge s D es ig na te p er so n fro m th e la bo ra to ry w ho w ill w or k w ith p ub lic h ea lth au th or iti es to a ss es s c lu st er s a nd b e pa rt o f o ut br ea k re sp on se te am Be tw ee n th e la bo ra to ry a nd p ub lic he al th a ut ho rit y, d ev el op c rit er ia fo r de te ct in g cl us te rs D oc um en t t he m et ho ds fo r c lu st er de te ct io n in a s ta nd ar d op er at in g pr oc ed ur e En su re th er e is m ul tis ec to ra l co lla bo ra tio n w he n us in g W G S fo r ou tb re ak in ve st ig at io ns 61 Web Annex O. Template for managing implementation Ac tiv iti es Ta sk s Pe rs on re sp on si bl e Ye ar Ye ar Ja n Fe b M ar Ap r M ay Ju n Ju l Au g Se p O ct N ov D ec Ja n Fe b M ar Ap r M ay Ju n Ju l Au g Se p O ct N ov D ec  H um an re so ur ce s D efi ne st affi ng re qu ire m en ts fo r o p- er at io n of th e ou tb re ak in ve st ig at io n sy st em D et er m in e w he th er n ew st aff n ee d to be re cr ui te d Id en tif y tra in in g op tio ns fo r u ps ki llin g ex ist in g st aff  M ea su ri ng s ys te m p er fo rm an ce D efi ne w he n th e ro le o f W G S in th e ou tb re ak in ve st ig at io n sy st em w ill be ev al ua te d D ec id e on th e ou tb re ak in ve st ig at io n at tri bu te s t ha t w ill b e im po rt an t t o ev al ua te Co m pl et e th e de sc rip tio n of th e fo od bo rn e ou tb re ak in ve st ig at io n sy st em Tr an si tio ni ng fr om tr ad iti on al ty pi ng m et ho ds (i f a pp lic ab le ) D efi ne th e pe rio d of tr an sit io n w he re tra di tio na l t yp in g w ill ru n in p ar al le l w ith W G S Pr ov id e a st at em en t a bo ut h ow tra ns iti on w ill be m an ag ed D efi ne h ow tr ad iti on al ty pi ng re su lts w ill be co m pa re d w ith W G S re su lts Co m pl et e th e su rv ei lla nc e sy st em de sc rip tio n 62 Whole genome sequencing as a tool to strengthen foodborne disease surveillance and response. Module 3. Whole genome sequencing in foodborne disease routine surveillance. Web annexes Ac tiv iti es Ta sk s Pe rs on re sp on si bl e Ye ar Ye ar Ja n Fe b M ar Ap r M ay Ju n Ju l Au g Se p O ct N ov D ec Ja n Fe b M ar Ap r M ay Ju n Ju l Au g Se p O ct N ov D ec D ev el op a b us in es s ca se Fi nd e xa m pl es o f t he b en efi ts o f W G S in fo od bo rn e di se as e su rv ei lla nc e Es tim at e co st o f W G S ac co rd in g to th e fo od bo rn e ou tb re ak in ve st ig at io n sy st em d es cr ip tio n Co nd uc t a co st –b en efi t a na ly sis to in cl ud e in th e bu sin es s c as e Co m pl et e bu sin es s c as e Su bm it bu sin es s c as e to se ni or m an ag em en t f or a pp ro va l a nd fu nd in g Co m m un ic at io n En su re th er e is a m ec ha ni sm fo r up da tin g se ni or d ec isi on -m ak er s Pi lo t s tu dy Se t u p an a dv iso ry g ro up fo r t he p ilo t st ud y O ut lin e w ho w ill be in vo lv ed in th e pi lo t s tu dy te am D efi ne o bj ec tiv es o f t he p ilo t s tu dy D efi ne p ilo t s tu dy q ue st io ns D efi ne sc op e of th e pi lo t s tu dy , in lin e w ith th e su rv ei lla nc e sy st em de sc rip tio n D ec id e if pi lo t w ill b e re tro sp ec tiv e or pr os pe ct iv e O ut lin e m et ho do lo gy to b e us ed , ba se d on th e su rv ei lla nc e sy st em de sc rip tio n If de te rm in in g or v al id at in g in te rp re ta tio n cr ite ria , s pe ci fy h ow it w ill be d on e 63 Web Annex O. Template for managing implementation Ac tiv iti es Ta sk s Pe rs on re sp on si bl e Ye ar Ye ar Ja n Fe b M ar Ap r M ay Ju n Ju l Au g Se p O ct N ov D ec Ja n Fe b M ar Ap r M ay Ju n Ju l Au g Se p O ct N ov D ec Pi lo t s tu dy Li st th e ke y m ile st on es Li st th e de liv er ab le s In cl ud e co m m un ic at io ns in th e pi lo t st ud y pl an D efi ne th e ev al ua tio n cr ite ria a nd h ow th ey w ill be m ea su re d Cr ea te a lo g to sy st em at ic al ly do cu m en t b ar rie rs /c on st ra in ts d ur in g th e pi lo t s tu dy Co m pl et e pi lo t s tu dy p la n Im pl em en t p ilo t s tu dy Ev al ua te p ilo t s tu dy a nd w rit e a re po rt Fu ll- sc al e im pl em en ta tio n M ak e ne ce ss ar y ad ju st m en ts to th e ou tb re ak in ve st ig at io n sy st em de sc rip tio n in lig ht o f t he e va lu at io n of th e pi lo t s tu dy Co m m en ce fu ll- sc al e im pl em en ta tio n 64 Whole genome sequencing as a tool to strengthen foodborne disease surveillance and response. Module 3. Whole genome sequencing in foodborne disease routine surveillance. Web annexes

Nutrition and Food Safety Department World Health Organization 20 avenue Appia 1211 Geneva Switzerland Email: nfs@who.int Website: https://www.wh o.int/teams/nutrition-and-food-safety/about