Big-data scheme for inquiring the shared mobilome in livestock and human gut microbiomes
Author(s): Shivang Bhanushali,Tuomas Borman,Katariina Pärnänen,Leo M Lahti
Affiliation(s): Department of Computing, University of Turku.
Background: Livestock farms serve as focal points for the emergence and dissemination of Antibiotic Resistance Genes (ARGs) due to constricted space compared to the volume of livestock and extensive antibiotic use. Nevertheless, only a handful of studies have conducted a thorough evaluation of the spread of antibiotic resistance originating from food systems, especially poultry farms which promote zoonosis. Moreover, Mobile genetic elements (MGEs) that play a fundamental role in the global spread of antimicrobial resistance (AMR) have often been overlooked in strategies to combat this crisis. Beyond their role as mere carriers, MGEs undergo unique evolutionary trajectories and face distinct selection pressures compared to their host cells which makes MGEs an important group of resistance determinants. In this context, we propose that ARGs and mobile elements, upon selection within the livestock gut, could cross the host barrier and elevate the risk of human exposure to antimicrobial-resistant infections. Objective: In support of our hypothesis, we seek to investigate the shared mobilome within the openly available human and livestock gut microbiome datasets, identifying shared features between the two hosts and further exploring the geographical overlap between the datasets. Methods: Our custom approach will be searching and retrieving shotgun metagenomic assemblies from the MGnify platform using the MGnifyR package; part of miaverse microbiome analysis ecosystem. For improved resolution, we place a considerable emphasis on prioritising data from long-read sequencing technologies. Using metagenomic assemblies as the input, we plan to implement mobilome annotation pipeline; a Nextflow-based wrapper designed to detect mobile genetic elements in prokaryotic genomes and metagenomes. The pipeline handles the query at various levels starting from preprocessing, prediction, annotation, integration and postprocessing. For downstream analysis, we will incorporate R and Bioconductor ecosystems; miaverse for microbiome data analysis, alongside TreeSummarizedExperiment and MultiAssayExperiment data containers. This integration ensures data harmonization while maintaining the portability and reproducibility of our workflow. Conclusion: A shift in our approach to addressing AMR could be facilitated by reevaluating our understanding of these mobile elements. Therefore, focusing on MGEs rather than exclusively on host-centric perspectives holds promise in effectively tackling the resistance crisis. By leveraging big-data analysis and the high resolution of the applied computation tools, the developed pipeline offers a systematic means to investigate antibiotic resistance dispersion.