Detection of pathogens and antimicrobial resistant genes from urine within 5 hours using Nanopore sequencing

Abstract

Purpose: Urinary Tract Infection (UTI) is a prevalent global health concern accounting for 1-3% of primary healthcare visits. The current methods for UTI diagnosis have a high turnaround time of 3-5 days for pathogen identification and susceptibility testing. This work is a proof-of-concept study aimed at determining the detection limit by establishing a culture and amplification-free DNA extraction methodology from spiked urine samples followed by real-time Nanopore sequencing and data analysis. Methods: This study first establishes an optical density culture-based method for spiking healthy urine samples with the six most prevalent uropathogens. Pathogens were spiked at two clinically significant concentrations of 10^3 and 10^5 CFU/ml. Three commercial DNA extraction kits were investigated based on the quantity of isolated DNA, average processing time, elution volume and the average cost incurred per extraction. The outperforming kit was used for direct DNA extraction and subsequent sequencing on MinION and Flongle flowcells. Results: The Blood and Tissue kit outperformed the other kits. All pathogens were identified at a concentration of 10^5 CFU/ml within ten minutes, and the corresponding AMR genes were detected within three hours of the sequencing start. The overall turnaround time including the DNA extraction and sequencing steps was five hours. Moreover, we also demonstrate that the identification of some pathogens and antibiotic-resistance genes was possible at a spike concentration of 10^3 CFU/mL. Conclusion: This study shows great promise toward reducing the time required for making an informed antibiotic administration from approximately 48 hours to five hours thereby reducing the number of empirical doses and saving lives.

Real-time Taxonomic Characterization of Long-read Mixed-species Sequencing Samples in Sorted Motif Distance Space:Voyager

“…We collected datasets from all available ONT sequencing platforms, flowcells, and relevant sample types to establish expected performance in clinical, animal, and limited environmental community samples, including blood infection data obtained using MinION 14 (ENA PRJEB60525 ) and Flongle 15 (ENA PRJEB49072 ), urine infection data 16 using MinION (ENA PRJEB73819 ) milk mastitis data 2 using MinION (ENA PRJEB53532 ), and environmental mock community data 17 using GridION and PromethION (ENA PRJEB29504). Sample files containing low amounts of data (< 100,000 reads) were excluded.…”