PDF(Pubmed)
Abstract：
Oxford Nanopore sequencing is one of the high-throughput sequencing technologies that facilitates the reconstruction of metagenome-assembled genomes (MAGs). This study aimed to assess the potential of long-read assembly algorithms in Oxford Nanopore sequencing to enhance the MAG-based identification of bacterial pathogens using both simulated and mock communities. Simulated communities were generated to mimic those on fresh spinach and in surface water. Long reads were produced using R9.4.1+SQK-LSK109 and R10.4 + SQK-LSK112, with 0.5, 1, and 2 million reads. The simulated bacterial communities included multidrug-resistant Salmonella enterica serotypes Heidelberg, Montevideo, and Typhimurium in the fresh spinach community individually or in combination, as well as multidrug-resistant Pseudomonas aeruginosa in the surface water community. Real data sets of the ZymoBIOMICS HMW DNA Standard were also studied. A bioinformatic pipeline (MAGenie, freely available at https://github.com/jackchen129/MAGenie) that combines metagenome assembly, taxonomic classification, and sequence extraction was developed to reconstruct draft MAGs from metagenome assemblies. Five assemblers were evaluated based on a series of genomic analyses. Overall, Flye outperformed the other assemblers, followed by Shasta, Raven, and Unicycler, while Canu performed least effectively. In some instances, the extracted sequences resulted in draft MAGs and provided the locations and structures of antimicrobial resistance genes and mobile genetic elements. Our study showcases the viability of utilizing the extracted sequences for precise phylogenetic inference, as demonstrated by the consistent alignment of phylogenetic topology between the reference genome and the extracted sequences. R9.4.1+SQK-LSK109 was more effective in most cases than R10.4+SQK-LSK112, and greater sequencing depths generally led to more accurate results.IMPORTANCEBy examining diverse bacterial communities, particularly those housing multiple Salmonella enterica serotypes, this study holds significance in uncovering the potential of long-read assembly algorithms to improve metagenome-assembled genome (MAG)-based pathogen identification through Oxford Nanopore sequencing. Our research demonstrates that long-read assembly stands out as a promising avenue for boosting precision in MAG-based pathogen identification, thus advancing the development of more robust surveillance measures. The findings also support ongoing endeavors to fine-tune a bioinformatic pipeline for accurate pathogen identification within complex metagenomic samples.
摘要：
牛津纳米孔测序是促进宏基因组组装基因组（MAG）重建的高通量测序技术之一。这项研究旨在评估长读组装算法在牛津纳米孔测序中的潜力，以使用模拟和模拟社区增强基于MAG的细菌病原体鉴定。生成模拟群落以模拟新鲜菠菜和地表水中的群落。使用R9.4.1+SQK-LSK109和R10.4+SQK-LSK112产生长读数,具有0.5、1和2百万个读数。模拟的细菌群落包括多重耐药的肠道沙门氏菌血清型海德堡，蒙得维的亚,和新鲜菠菜群落中的鼠伤寒菌单独或组合，以及地表水群落中的多重耐药铜绿假单胞菌。还研究了ZymoBIOMICSHMWDNA标准的真实数据集。生物信息学管道(MAGenie，免费提供在https://github.com/jackchen129/MAGenie）结合宏基因组组装，分类学分类，并开发了序列提取来重建宏基因组组装中的MAG草案。基于一系列基因组分析评估了五个组装者。总的来说,弗莱的表现优于其他装配商，紧随其后的是沙斯塔,Raven,还有Uniculer,而Canu的表现最差。在某些情况下,提取的序列产生了MAG草案,并提供了抗菌素抗性基因和可移动遗传元件的位置和结构.我们的研究展示了利用提取的序列进行精确的系统发育推断的可行性，正如参考基因组和提取序列之间的系统发生拓扑结构的一致比对所证明的那样。在大多数情况下,R9.4.1+SQK-LSK109比R10.4+SQK-LSK112更有效,并且更大的测序深度通常导致更准确的结果。重要性通过检查不同的细菌群落，特别是那些拥有多种肠道沙门氏菌血清型的人，这项研究对于揭示长读数组装算法通过牛津纳米孔测序改善基于宏基因组组装基因组(MAG)的病原体鉴定的潜力具有重要意义.我们的研究表明，长阅读组装是提高基于MAG的病原体鉴定精度的有希望的途径，从而推进更强有力的监测措施的发展。这些发现还支持正在进行的努力，以微调生物信息学管道，以在复杂的宏基因组样品中进行准确的病原体鉴定。
