Herbert Josephine, Thompson Stanley, Beckett Angela H, Robson Samuel C
Centre for Enzyme Innovation, University of Portsmouth, Portsmouth, Hampshire, PO1 2DT, UK.
Institute of Life Sciences and Healthcare, University of Portsmouth, Portsmouth, Hampshire, PO1 2DT, UK.
Environ Microbiome. 2025 May 5;20(1):47. doi: 10.1186/s40793-025-00704-7.
Metagenomics, the genomic analysis of all species present within a mixed population, is an important tool used for the exploration of microbiomes in clinical and environmental microbiology. Whilst the development of next-generation sequencing, and more recently third generation long-read approaches such as nanopore sequencing, have greatly advanced the study of metagenomics, recovery of unbiased material from microbial populations remains challenging. One promising advancement in genomic sequencing from Oxford Nanopore Technologies (ONT) is adaptive sampling, which enables real-time enrichment or depletion of target sequences. As sequencing technologies continue to develop, and advances such as adaptive sampling become common techniques within the microbiological toolkit, it is essential to evaluate the benefits of such advancements to metagenomic studies, and the impact of methodological choices on research outcomes.
Given the rapid development of sequencing tools and chemistry, this study aimed to demonstrate the impacts of choice of DNA extraction kit and sequencing chemistry on downstream metagenomic analyses. We first explored the quality and accuracy of 16S rRNA amplicon sequencing for DNA extracted from the ZymoBIOMICS Microbial Community Standard, using a range of commercially available DNA extraction kits to understand the effects of different kit biases on assessment of microbiome composition. We next compared the quality and accuracy of metagenomic analyses for two nanopore-based ligation chemistry kits with differing levels of base-calling error; the older and more error-prone (~ 97% accuracy) LSK109 chemistry, and newer more accurate (~ 99% accuracy) LSK112 Q20 + chemistry. Finally, we assessed the impact of the nanopore sequencing chemistry version on the output of the novel adaptive sampling approach for real-time enrichment of the genome for the yeast Saccharomyces cerevisiae from the microbial community.
Firstly, DNA extraction kit methodology impacted the composition of the yield, with mechanical bead-beating methodologies providing the least biased picture due to efficient lysis of Gram-positive microbes present in the community standard, with differences in bead-beating methodologies also producing variation in composition. Secondly, whilst use of the Q20 + nanopore sequencing kit chemistry improved the base-calling data quality, the resulting metagenomic assemblies were not significantly improved based on common metrics and assembly statistics. Most importantly, we demonstrated the effective application of adaptive sampling for enriching a low-abundance genome within a metagenomic sample. This resulted in a 5-7-fold increase in target enrichment compared to non-adaptive sequencing, despite a reduction in overall sequencing throughput due to strand-rejection processes. Interestingly, no significant differences in adaptive sampling enrichment efficiency were observed between the older and newer ONT sequencing chemistries, suggesting that adaptive sampling performs consistently across different library preparation kits.
Our findings underscore the importance of selecting a DNA extraction methodology that minimises bias to ensure an accurate representation of microbial diversity in metagenomic studies. Additionally, despite the improved base-calling accuracy provided by newer Q20 + sequencing chemistry, we demonstrate that even older ONT sequencing chemistries can achieve reliable metagenomic sequencing results, enabling researchers to confidently use these approaches depending on their specific experimental needs. Critically, we highlight the significant potential of ONT's adaptive sampling technology for targeted enrichment of specific genomes within metagenomic samples. This approach offers broad applicability for enriching target organisms or genetic elements (e.g., pathogens or plasmids) or depleting unwanted DNA (e.g., host DNA) in diverse sample types from environmental and clinical studies. However, researchers should carefully weigh the benefits of adaptive sampling against the potential trade-offs in sequencing throughput, particularly for low-abundance targets, where strand rejection can lead to pore blocking. These results provide valuable guidance for optimising adaptive sampling in metagenomic workflows to achieve specific research objectives.
宏基因组学是对混合群体中所有物种进行基因组分析，是临床和环境微生物学中用于探索微生物群落的重要工具。尽管下一代测序技术以及最近的第三代长读长方法（如纳米孔测序）的发展极大地推动了宏基因组学研究，但从微生物群体中获取无偏差材料仍然具有挑战性。牛津纳米孔技术公司（ONT）在基因组测序方面的一项有前景的进展是自适应采样，它能够实时富集或去除目标序列。随着测序技术不断发展，诸如自适应采样等进展成为微生物学工具包中的常用技术，评估这些进展对宏基因组学研究的益处以及方法选择对研究结果的影响至关重要。
鉴于测序工具和化学方法的快速发展，本研究旨在证明DNA提取试剂盒和测序化学方法的选择对下游宏基因组分析的影响。我们首先使用一系列市售DNA提取试剂盒，探索从ZymoBIOMICS微生物群落标准品中提取的DNA进行16S rRNA扩增子测序的质量和准确性，以了解不同试剂盒偏差对微生物群落组成评估的影响。接下来，我们比较了两种基于纳米孔的连接化学试剂盒（碱基识别错误水平不同）进行宏基因组分析的质量和准确性；较旧且错误率较高（约97%准确性）的LSK109化学方法，以及更新且更准确（约99%准确性）的LSK112 Q20 +化学方法。最后，我们评估了纳米孔测序化学版本对从微生物群落中实时富集酿酒酵母基因组的新型自适应采样方法输出的影响。
首先，DNA提取试剂盒方法影响了产量的组成，机械珠磨方法由于能有效裂解群落标准品中存在的革兰氏阳性微生物，提供的偏差最小，且珠磨方法的差异也会导致组成的变化。其次，虽然使用Q20 +纳米孔测序试剂盒化学方法提高了碱基识别数据质量，但基于常见指标和组装统计，所得宏基因组组装并未得到显著改善。最重要的是，我们证明了自适应采样在宏基因组样本中富集低丰度基因组的有效应用。与非自适应测序相比，这导致目标富集增加了5至7倍，尽管由于链排斥过程总体测序通量有所降低。有趣的是，在较旧和较新的ONT测序化学方法之间未观察到自适应采样富集效率的显著差异，这表明自适应采样在不同文库制备试剂盒中表现一致。
我们的研究结果强调了选择一种能将偏差降至最低的DNA提取方法的重要性，以确保在宏基因组学研究中准确呈现微生物多样性。此外，尽管更新的Q20 +测序化学方法提高了碱基识别准确性，但我们证明即使是较旧的ONT测序化学方法也能获得可靠的宏基因组测序结果，并使研究人员能够根据其特定实验需求自信地使用这些方法。至关重要的是，我们强调了ONT的自适应采样技术在宏基因组样本中靶向富集特定基因组的巨大潜力。这种方法在环境和临床研究的各种样本类型中，对于富集目标生物体或遗传元件（如病原体或质粒）或去除不需要的DNA（如宿主DNA）具有广泛的适用性。然而，研究人员应仔细权衡自适应采样的益处与测序通量方面潜在的权衡，特别是对于低丰度目标，链排斥可能导致孔堵塞。这些结果为优化宏基因组工作流程中的自适应采样以实现特定研究目标提供了有价值的指导。
