Comprehensive profiling of antibiotic resistance genes in diverse environments and novel function discovery

Antibiotic resistance genes (ARGs) have emerged in pathogens and are arousing worldwide concern, and accurately identifying unknown ARGs is a formidable challenge in studying the generation and spread of antibiotic resistance in diverse environments. Current methods can identify known ARGs but have limited utility for the discovery of novel ARGs, thus rendering the profiling of ARGs incomprehensive. Here, we developed ONN4ARG, an ontology-aware deep learning approach for comprehensive ARG discovery. Systematic evaluation revealed that ONN4ARG outperforms previous methods in terms of efficiency, accuracy, and comprehensiveness. Experiments using 200 million microbial genes collected from 815 metagenomic samples from diverse environments or hosts have resulted in 120,726 candidate ARGs, of which more than 20% are not yet present in public databases. The comprehensive set of ARGs revealed environment-specific and host-specific patterns. The wet-lab functional validation together with structural investigation have validated a novel streptomycin resistance gene from oral microbiome samples, confirming ONN4ARG��s ability to discover novel functions. In summary, ONN4ARG enables comprehensive ARG discovery toward a grand view of ARGs worldwide.