The ENGAGE Project conducts benchmarking exercises to provide resources and guidance for next generation sequencing.
In the capacity to provide resources and guidance for next generation sequencing, the EFSA project ENGAGE conducts quarterly benchmarking exercises – some smaller and some larger.
In the last quarter, the aim was to test bioinformatic serotyping tools: Public Health England’s MOST and a command line version of SeqSero by DTU Food. A total of 798 strains compressed by 137 different serotypes were tested.
The results clearly demonstrate that serotyping using NGS data is a very feasible option. The best tools revealed an 84% correlation with the conventional serotyping, and this is a conservative number, considering that none of the isolates have been retested to ensure correct serotyping.
The plan is to expand the exercise by including the tool SISTR.
The results of the analysis are available from ENGAGE web site.
It is important to conduct benchmarking as this will allow the users to evaluate the performance and the tool that fit best to the purpose of the analysis.
For this specific benchmarking exercise, an additional objective was to initiate a change to the EU regulations for isolation and identification of Salmonella; the regulation indicates that Salmonella isolates need to be serotyped by conventional methodologies such as slide agglutination. The requirement hampers institutes that are changing their monitoring from conventional microbiology to an NGS approach.
A second serotyping benchmarking exercise will be scheduled for the first or second quarter of 2017, aiming more toward the requirements of the ISO 16140-2:2016 “Microbiology of the food chain -- Method validation: Protocol for the validation of alternative (proprietary) methods against a reference method”.
More information is available via the ENGAGE web site.