对不同公司16S rRNA基因MiSeq测序数据的一致性分析

Consistency analysis of MiSeq sequencing data of 16S rRNA genes from different biotech companies

摘要:

【背景】高通量测序技术已经广泛应用于环境微生物研究的各个领域。不同原理的测序平台以及众多生物公司的出现为各个科研团队提供了各具特色的测序技术支持，在满足了不同研究需要的同时，也产生了多种多样的测序数据。这些基于不同测序平台，以及同一测序平台下不同测序公司所产生的数据之间是否具有通用性，一直以来都是广大科研学者所关注的。【目的】探究同一样品在基于MiSeq测序平台下，不同测序环境以及不同测序深度对实验数据的影响，并进一步探究造成差异的原因，以及这些差异对实验结果的影响。【方法】从鄱阳湖松门山、南矶山、饶河、白沙洲采集底泥沉积物样品，分别在2个公司进行不同测序深度16S rRNA基因V3?V4区高通量测序，并比较分析2组测序数据。【结果】2组数据反映的微生物群落结构在实验样地间的分布规律具有高度的相似性，但稀有微生物的差异导致他们在PCoA以及聚类分析中被分为两簇。关系网络关联分析发现具有较高测序深度的B组数据反映了更为复杂的微生物间相互作用，部分稀有微生物如Deferribacteres (脱铁杆菌门)、Lentisphaerae (黏胶球形菌门)等在群落中发挥着重要的作用。METAGENassist功能预测发现了他们在Atrazine metabolism、Chitin degradation、Sulfate reducer、Nitrogen fixation等14类功能上存在差异。【结论】不同的测序环境对实验数据造成的影响可以通过数据质控过程减弱甚至排除，而测序深度的不同则会对测序数据产生显著影响。这种影响主要体现在较深的测序深度会显著增加稀有微生物的丰富度，进而有利于增强我们对环境微生物群落整体功能的认识。

Abstract:

[Background] High-throughput sequencing technology has been widely used in the research field of environmental microbiology. Due to sequencing platform based on different principles, and personalized service biotech companies provided, huge amounts of various sequencing data are emerged. Although personalized service is good to meet customers’ different requirements, there is a widespread concern if the sequencing data from different sequencing platforms or different companies could be equally treated. [Objective] The aim of this study is to explore the impacts of different sequencing conditions and sequencing depths on the final sequencing data of the same sample using MiSeq sequencing platform, and further to find out the reasons for the differences, and the subsequent effects of these differences on the experimental data. [Methods] Sediment samples were collected from Songmenshan Region, Nanjishan, Raohe River and Baishazhou of the Poyang Lake. High-throughput sequencing of 16S rRNA gene V3?V4 region was performed in two biotech companies with different sequencing depths, and two sets of data were compared. [Results] Two sets of data showed highly similarity in microbial community structure, but the abundance difference between the rare species resulted in a different pattern in the PCoA and cluster analysis. Co-occurrence network revealed that the data with higher sequencing depth could reflect more complex interactions between and within microbial taxa. Some rare species such as Deferribacteres and Lentisphaerae were found to be important for the community eco-function. A total of 14 categories of differentiated metabolism were found between two datasets by METAGEN assist functional forecast method, including Atrazine metabolism, Chitin degradation, Sulfate reducer, Nitrogen fixation and so on. [Conclusion] The impacts of different sequencing environments on experimental data can be reduced or even eliminated by data quality control processes, but different sequencing depths have a significant impact on the sequencing data. Increasing the sequencing depth can significantly improve the richness of rare species, and thus supply a comprehensive knowledge of microbial community function.