Comparative Performance of Portable DNA Extraction Protocols and Bioinformatics Workflows for Rapid Detection of Pathogens and Antimicrobial Resistance Using Oxford Nanopore Sequencing

Abstract

ABSTRACT Oxford Nanopore technology (ONT) enables rapid, portable pathogen identification and detection of antimicrobial resistance (AMR). Still, the reliability of downstream genomic analyses is highly dependent on DNA extraction quality, particularly in resource-limited settings. This study comparatively evaluated four portable bacterial DNA extraction protocols, derived from three commercial kits, to determine their impact on nanopore sequencing performance, bioinformatics workflow completion, and field deployability. Six Gram-negative bacterial isolates ( Escherichia coli , n= 4; Pseudomonas sp., n= 1; and Salmonella sp., n= 1) were processed using four extraction protocols: SwiftX DNA, SwiftX DNA with proteinase K (ProtK), SwiftX ParaBact, and NucleoSpin Microbial. Twenty-four DNA extracts (6 isolates x 4 protocols) were sequenced on a single multiplexed MinION R10.4.1 flow cell. Sequencing data were analysed using validated Galaxy-based generic and species-specific pipelines, with workflow completion defined as successful progression through quality control, assembly, virulence, plasmid and AMR detection modules. DNA purity varied substantially by extraction protocol and was strongly associated with workflow success. NucleoSpin Microbial achieved 100% workflow completion, SwiftX ParaBact achieved 83%, while both SwiftX DNA-based protocols failed to complete full workflows. Higher A260/A280 ratios were strongly correlated with successful workflow completion (Spearman’s ρ = 0.767, p < 0.0001). Importantly, key AMR genes required to classify isolates as multidrug resistant were consistently detected using both NucleoSpin Microbial and SwiftX ParaBact extractions. However, NucleoSpin Microbial assemblies showed significantly higher contiguity and enabled broader and more complete detection of virulence factors, pathogenicity islands, plasmid replicons, and accessory AMR genes, reflecting enhanced genomic resolution. IMPORTANCE Rapid whole-genome sequencing is increasingly used to detect antimicrobial resistance and guide public health responses, but its reliability depends strongly on how bacterial DNA is extracted. In this study, we show that DNA extraction method choice has a major impact on Oxford Nanopore sequencing performance across clinically relevant bacteria. While silica-column-based extraction maximised genomic completeness and analytical depth, paramagnetic bead-based reverse purification offered superior portability with sufficient resolution for frontline AMR surveillance. These findings highlight a practical trade-off between field deployability and high-resolution genomic characterisation in low-resource settings.