L-EasyARG: A long-read metagenomics tool for rapid antibiotic-resistance-gene profiling

l-EasyARG is a pipeline for long-read metagenomic profiling of antibiotic resistance genes (ARGs) and their genetic context. It integrates ARG identification, plasmid and mobile genetic element (MGE) detection, taxonomic classification, and further identifies ARGs carried by human bacterial pathogens (HBPs) defined in the WHO and ESKAPE pathogen lists. Benchmarking on simulated datasets with error rates ranging from 0% to 15% demonstrates optimal F1-scores at ≤5% errors, outperforming existing tools such as ARGpore2 and DeepARG. Application to hospital wastewater samples validates its ability to capture disinfection-induced shifts in ARG composition, ARG-MGE co-occurrence, and ARG-host associations. l-EasyARG generates comprehensive outputs including abundance profiles, co-occurrence networks, and interactive visualizations, enabling efficient resistome surveillance and risk assessment. The pipeline is user-friendly and locally executable.