A metagenomic survey of global antimicrobial resistance

Abstract

Antimicrobial resistance (AMR) poses a major global health threat, yet understanding of environmental reservoirs remains limited. This study analysed 1,758 metagenomes to assess microbial species and antimicrobial resistance gene (ARG) diversity across global soil and aquatic environments. Using Kraken2 for taxonomic classification and DeepARG for ARG identification, coupled with diversity analyses, this research revealed distinct patterns in microbial and antibiotic resistance gene profiles. Microbial communities showed similar overall diversity between soil and aquatic environments (Shannon indices: 5.409 and 5.410 respectively) but exhibited distinct structural patterns. Urban wastewater demonstrated the highest microbial diversity in aquatic settings (Shannon index: 5.43), while cold biomes showed highest diversity in soils (Shannon index: 5.63). ARG profiles revealed different patterns, with soil environments showing higher median diversity (4.68) compared to aquatic environments (4.52). Notably, anthropogenic soils exhibited the highest ARG diversity (Shannon index: 5.71) despite reduced microbial diversity, highlighting human activity's role in ARG proliferation. The soil core resistome was substantially larger than aquatic (179 vs 113 ARGs), with 101 shared ARGs between environments. ESKAPE pathogens were most prevalent in human-impacted environments, reaching 8.32% in groundwater and 8.23% in temperate soils. This comprehensive analysis reveals how environmental conditions and human activities distinctly shape microbial communities and their associated resistomes. These findings underscore the urgent need for integrated AMR surveillance beyond clinical settings and highlight critical intervention points for mitigating the spread of antibiotic resistance, particularly in agricultural and urban environments where human activities substantially influence resistance patterns.