Oxford Nanopore sequencing of Escherichia coli ST131

We evaluated the performance of several popular tools for each step of genome reconstruction, including base-calling, filtering, assembly, and polishing. We also assessed overall genome accuracy using ONT both natively and with Illumina. All steps were validated using the high-quality complete reference genome for the Escherichia coli sequence type (ST)131 strain EC958. Software chosen at each stage were incorporated into our final pipeline, MicroPIPE.Further validation of MicroPIPE was carried out using 11 additional ST131 E. coli isolates, which demonstrated that complete circularised chromosomes and plasmids could be achieved without manual intervention. Eight publicly available Gram-negative and Gram-positive bacterial isolates (with available raw ONT data and matched complete genomes) were also assembled using MicroPIPE. We found that revised basecalling and updated assembly of the majority of these genomes resulted in improved accuracy compared to the current publicly available complete genomes.

我们针对基因组重构的各个环节，包括碱基识别（base-calling）、序列过滤、基因组组装与序列抛光（polishing），对多款主流工具的性能开展了评估。同时采用原生牛津纳米孔（ONT）测序数据方案，以及结合因美纳（Illumina）测序数据的混合方案，评估了基因组的整体准确性。所有流程均通过大肠杆菌（Escherichia coli）序列型（ST）131菌株EC958的高质量完整参考基因组完成了验证。各环节筛选出的最优软件被整合至我们的最终分析流程MicroPIPE中。我们额外选取11株ST131型大肠杆菌分离株对MicroPIPE开展了进一步验证，结果表明该流程可在无需人工干预的情况下，获得完整的环形染色体与质粒序列。我们还采用MicroPIPE对8株公开可获取的革兰氏阴性（Gram-negative）与革兰氏阳性（Gram-positive）细菌分离株（配有原始ONT测序数据及匹配的完整参考基因组）进行了组装。研究发现，相较于当前公开的完整基因组，对这批基因组中的绝大多数采用优化后的碱基识别流程与更新后的组装算法后，其序列准确性得到了显著提升。