EMtrace: offline nanopore metagenomics on a laptop for rapid viral diagnostics, genome recovery and 16S profiling

EMtrace, an offline laptop pipeline for nanopore metagenomics, achieved the highest species-level F1 scores on SARS-CoV-2 clinical samples (0.97) with only one false-positive species.

• Why it matters: Rapid nanopore read generation is not matched by metagenomic analysis in settings with limited internet, sensitive data, or modest compute, especially for viral diagnostics and profiling.
• What they did: The pipeline classifies MinKNOW reads, estimates taxon abundance using expectation-maximization, and assembles reference-based consensus genomes; it was benchmarked on simulated, mock-community, and clinical datasets.
• The result: In replayed runs it detected the target virus within the first 4,000-read batch (5–13 min), and with a shared 16S reference it reported two to three times fewer false-positive species than Emu.

AI summary of the abstract · may contain errors · not medical advice · read the paper.