The sample analysis is based on several levels of analysis to optimise power of the study, with three levels which vary in their depth of analysis and their applicability to large studies. While metagenomics provides an excellent coverage of resistance determinants in the total community for each sample, it is not feasible within the framework of the project to analyse thousands of samples despite the fact that the costs for metagenomic analyses have come down considerably. The following levels of analysis have been identified:
A subset of the samples (a random cross-section of approximately 700 samples from high risk primary production sectors – animals and humans) will be analysed using metagenomic approaches to give insight in the complete spectrum of resistance genes and mobile genetic elements present. This step should lead to adequate identification of relevant genes in samples from animals, humans and the environment. Metagenomics will be applied to samples from the whole microbial community, with the advantage that not only the resistance determinants in the small subset of commonly isolated bacteria, but in the total community will be included. This might reveal the existence of new or emerging types of resistance currently not yet identified.
The most prevalent or relevant genes which have been identified by metagenomics, and which appear to be present reproducibly, will be analysed by the use of qPCR on the complete sample set of approximately 10 000 samples including, humans and environmental samples. This will give additional information to the metagenomic and array analysis in the sense that it gives information about the concentration of a particular gene in a sample.
Isolates of the sentinel organism E. coli retrieved from all countries and all samples will be analysed for their resistance pattern by determining minimum inhibitory concentrations (MICs) for a panel of antimicrobials, and a subset of the isolates will be sequenced, with a focus on isolates from the sentinel countries for the in-depth epidemiological study. These results will enable a pan-European analysis of resistance in relation to type of production industry, qualitative and quantitative use of antimicrobial agents, exploration of different transmission pathways, etc. By comparing resistance gene detection in selected strains and the whole microbial community, an insight into the value of present surveillance approaches will be obtained. Current surveillance systems are based on selection of sentinel organisms rather than the complete “resistome” as it circulates in the microbial community.