开源纳米孔测序流程用于克隆暴发分析,准确度达短读长水平
研究建立并优化了仅用纳米孔长读长测序的开源流程,用于克隆暴发流行病学分析。以10株无关铜绿假单胞菌优化建库和dorado分析参数后,在四组医院暴发菌株(肺炎克雷伯菌12株、铜绿假单胞菌11株、屎肠球菌10株、金黄色葡萄球菌10株)中验证,cgMLST最小生成树与Illumina短读长参考完全一致,全基因组水平每基因组仅少至2个不一致位点。
要点
- 优化dorado sup@v5.0.0碱基识别并加入纠错和细菌模型抛光可获得最佳性能。
- 四组暴发菌株的纳米孔cgMLST聚类与Illumina短读长参考完全一致。
- 全基因组水平与短读长组装高度一致,每基因组少至2个不一致位点。
核心结论
- 优化的建库与开源计算流程使仅用纳米孔测序即可进行克隆性和暴发分析,性能与Illumina短读长测序相当。
研究设计与样本
- 方法学研究,用10株无关临床铜绿假单胞菌优化流程,并在四组回顾性医院暴发菌株中验证。
主要发现
- 最佳性能为dorado sup@v5.0.0碱基识别联合dorado纠错和dorado polish细菌模型。
- 四组暴发菌株(肺炎克雷伯菌n=12、铜绿假单胞菌n=11、屎肠球菌n=10、金黄色葡萄球菌n=10)的纳米孔cgMLST最小生成树与Illumina短读长参考完全一致。
- 全基因组水平高度一致,与短读长组装相比每基因组少至2个不一致位点。
关键数值与推荐
- 测序数据分析使用dorado suite v0.9.1,包括碱基识别、组装前读长纠错和组装后抛光算法。
- 长读长组装和cgMLST分别使用Flye和pyMLST完成。
- 建库采用改良快速条形码策略并加入温度梯度以改善高GC含量基因组性能。
临床怎么用
- 该流程可用于医院感染控制中的暴发溯源和克隆性分析。
- 纳米孔测序成本更低、便携且快速,优化后准确度满足暴发调查需求。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
原文
UNLABELLED: In this work, we present an optimized nanopore long-read only sequencing workflow for epidemiologic analysis of clonal outbreaks built with open-source tools. A set of unrelated clinical Pseudomonas aeruginosa isolates (n = 10) was chosen for workflow optimization, and sequencing libraries were prepared using a modified rapid barcoding strategy that incorporates temperature ramps to improve performance for high-GC content genomes. Sequencing data were used to benchmark the performance of the dorado suite (v0.9.1), including its basecaller, pre-assembly read error correction, and post-assembly polishing algorithms. All long-read assemblies and core genome multilocus sequence typing (cgMLST) were performed with Flye and pyMLST, respectively. Results were compared with a standard reference Illumina short-read approach, and discordant positions were determined at the core and whole-genome levels. Optimal performance was found with dorado sup@v5.0.0 basecalling with the inclusion of dorado error correction and dorado polish with its bacterial model. This workflow was then validated with four retrospective hospital outbreak isolate sets, including Klebsiella pneumoniae (n = 12), P. aeruginosa (n = 11), Enterococcus faecium (n = 10), and Staphylococcus aureus (n = 10). The nanopore-only assemblies obtained from the optimized pipeline demonstrated fully concordant cgMLST-based minimum spanning trees compared to the Illumina short-read reference. At the whole-genome level, high concordance was also observed, with as few as two discordant positions per genome compared to short-read assemblies. This optimized library preparation and open-source computational workflow enables nanopore-only clonality and outbreak analysis with performance comparable to that of Illumina short-read sequencing and will contribute critically to hospital infection control. IMPORTANCE: For the past decade, bacterial whole-genome sequencing has been performed using high-accuracy short-read sequencing. More recently, long-read sequencing with Oxford Nanopore Technologies (ONT) instruments has emerged as a potential alternative based on multiple advantages, including lower costs, portability, and speed. However, this platform has suffered from basecall error rates that were too high for many applications in clinical microbiology, including outbreak tracing. With the release of new flow cell chemistries and basecall algorithms, the accuracy has improved dramatically, making this approach feasible for outbreak investigations. In this work, we optimize a streamlined nanopore-only workflow for epidemiologic analysis of bacterial pathogens. The workflow was validated with isolates from four previously identified clinical outbreaks with varying GC content and demonstrated fully concordant cgMLST clustering as compared to short-read references. This workflow will facilitate the broader implementation of ONT-only genomes and cgMLST analysis to assist in hospital outbreaks worldwide.
引用
- Vereecke N, Yoon TB, Luo TL, et al. An open-source nanopore-only sequencing workflow for analysis of clonal outbreaks delivers short-read level accuracy. J Clin Microbiol. 2025. PMID: 40679848. doi:10.1128/jcm.00664-25
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。