Methodological aspects of investigating the resistome in pig farm environments.

A typical One Health issue, antimicrobial resistance (AMR) development and its spread among people, animals, and the environment attracts significant research attention. The animal sector is one of the major contributors to the development and dissemination of AMR and accounts for more than 50 % of global antibiotics usage. The use of antibiotics exerts a selective pressure for resistant bacteria in the exposed microbiome, but many questions about the epidemiology of AMR in farm environments remain unanswered. This is connected to several methodological challenges and limitations, such as inconsistent sampling methods, complexity of farm environment samples and the lack of standardized protocols for sample collection, processing and bioinformatical analysis. In this project, we combined metagenomics and bioinformatics to optimise the methodology for reproducible research on the resistome in complex samples from the indoor farm environment. The work included optimizing sample collection, transportation, and storage, as well as DNA extraction, sequencing, and bioinformatic analysis, such as metagenome assembly and antibiotic resistance gene (ARG) detection. Our studies suggest that the current most optimal and cost-effective pipeline for ARG search should be based on Illumina sequencing of sock sample material at high depth (at least 25 M 250 bp PE for AMR gene families and 43 M for gene variants). We present a computational analysis utilizing MEGAHIT assembly to balance the identification of bacteria carrying ARGs with the potential loss of diversity and abundance of resistance genes. Our findings indicate that searching against multiple ARG databases is essential for detecting the highest diversity of ARGs.