Finally, Bulk Typing of Bacterial Species down to Strain Level using ON-rep-seq

Abstract

Despite the massive developments within culture-independent methods for detection and quantification of microorganisms during the last decade, culture-based methods remain a cornerstone in microbiology. We have developed a new method for bacterial DNA enrichment and tagmentation allowing fast (< 24h) and cost-effective species level identification and strain level differentiation using the MinION portable sequencing platform (ON-rep-seq). DNA library preparation takes less than 5h and ensures highly reproducible distribution of reads that can be used to generate strain level specific read length counts profiles (LCp). We have developed a pipeline that by correcting the random error of reads within peaks of LCp generates a set ([~]10 contigs per sample; 300bp - 3Kb) of high quality (>99%) consensus reads. Whereas, the information from high quality reads is used to retrieve species level taxonomy, comparison of LCp allows for strain level differentiation. With benchmarked 288 isolates identified on a single flow cell and a theoretical throughput to evaluate over 1000 isolates, our method allows for detailed bacterial identification for less than 2$ per sample at very high speed.