It is expected that WGS will become the sole standard method for genotyping of food and waterborne diseases (FWD) pathogens for public health purposes.
Laboratories that already perform WGS should still do current typing techniques so that data remain comparable across organisations and can be used for further validation.
Some residual capacity for current typing techniques should be kept in a few laboratories, to which others can send isolates if needed.
The expert opinion, produced by the FWD-NEXT Expert Group, focuses on Salmonella, Listeria monocytogenes, verocytotoxin–producing E. coli (VTEC) and Campylobacter.
It covers from sample provision and sequencing to data analysis and sharing and is for countries who are planning to, or are in the process of, implementing WGS for routine surveillance and outbreak investigation of FWD.
The paper produced 13 proposed future steps including assessing intra- and inter-laboratory reproducibility of WGS sequencing, explore whether a single global or EU/EEA-wide database can be established for each pathogen and develop software that can annotate (partially) assembled genomes with quality information per nucleotide.
An ISO standard on whole genome sequencing for typing and genomic characterisation of food products is also under development.
WGS cost
WGS can be less expensive than current typing methods for E. coli and Campylobacter. For Listeria, the cost is more or less the same, and for Salmonella, depending on throughput, it can be higher.
The total time required for WGS is already comparable to that of current typing methods. As WGS technology is evolving rapidly, the cost and total time can be expected to decrease further.
“Overall, the method is expected to become less expensive for all pathogens compared with current typing methods. Taking into account also the higher accuracy of the method for delineating epidemiologically relevant clusters, the potential for preventing additional cases through earlier detection is also higher than for current typing methods,” said ECDC.
Typing techniques currently supported by ECDC as validated and standardised methods for routine application are, depending on the pathogen, serotyping, multiple-locus variable-number tandem-repeat analysis (MLVA) and pulsed-field gel electrophoresis (PFGE).
Apart from the cost per isolate and total time, there are other factors when deciding to switch to WGS.
“Firstly, there is some cost associated with maintaining several different typing techniques per pathogen, in addition to the isolation and selected phenotyping methods that would still need to be performed,” said the report.
“Secondly, some of the current typing techniques are difficult and require a lot of experience, and laboratories may not have a sufficient number of qualified technicians for this task. As a result, maintaining capacity over time for these techniques may be an issue.”
While development of tools for analysis and interpretation of WGS data is rapidly progressing, there is much uncertainty and no consensus on translating them into practically useful information for public health purposes.
Culture independent diagnostic methods will require updating legislation on case definitions for EU level surveillance and likely also minimum requirements for providing samples for characterisation to national public health reference laboratories or ensuring decentralised characterisation.
Sampling frame
While defining a sampling frame is normally not an issue, in practice economical and technical factors usually determine how many samples are typed from which regions and with which timeliness, said the report.
The most realistic scenarios – at present and for the foreseeable future – are those where the primary laboratories provide pure cultures, or if not possible, clinical samples to the national reference laboratory, which then performs further typing.
The advent of WGS would not immediately change that, but in some cases hospitals are already setting up their own sequencing facilities. Private operators also exist already.
Obtaining the required amount of DNA is not expected to be a problem for most FWD pathogens as they can be easily cultured.
Sequencing platforms capable of producing very long reads may require more sophisticated extraction techniques to avoid DNA fragmentation.
Extraction is more difficult for gram-positive bacteria such as Listeria spp. than for Gram-negative bacteria such as Salmonella spp., E. coli and Campylobacter spp. due to the composition of the cell wall but an additional enzymatic step with lysozyme usually solves this problem.
Data storage
The amount of data generated per isolate by a sequencer is substantial. The raw read information consisting of sequence and individual base quality scores is typically stored as a FASTQ file.
As storage of the raw read data may well become a requirement for accreditation of the analysis pipeline, and likely also for litigation, it is very important to retain this information.
“Depending on the number of isolates that are sequenced, data storage may or may not be an issue, either technically or in terms of cost,” said the report.
“The sequencing and storage of, for example, 10,000 Salmonella raw reads files, a substantial amount at present, would require around 5 terabyte of uncompressed data storage space…Taking into account backup requirements, this number has to be doubled at least, and if the data are to be available efficiently at all times for analysis, it should at least be tripled.”