Xiao Fanshu, Yu Yuhe, Li Jinjin, Juneau Philippe, Yan Qingyun
Environmental Microbiomics Research Center, Guangdong Provincial Key Laboratory of Environmental Pollution Control and Remediation Technology, and the School of Environmental Science and Engineering, Sun Yat-sen University, Guangzhou, 510006, China.
Key Laboratory of Aquatic Biodiversity and Conservation of Chinese Academy of Sciences, Institute of Hydrobiology, Chinese Academy of Sciences, Wuhan, 430072, China.
Curr Microbiol. 2018 Sep;75(9):1240-1246. doi: 10.1007/s00284-018-1516-y. Epub 2018 May 25.
The 16S rRNA gene is one of the most commonly used molecular markers for estimating bacterial diversity during the past decades. However, there is no consistency about the sequencing depth (from thousand to millions of sequences per sample), and the clustering methods used to generate OTUs may also be different among studies. These inconsistent premises make effective comparisons among studies difficult or unreliable. This study aims to examine the necessary sequencing depth and clustering method that would be needed to ensure a stable diversity patterns for studying fish gut microbiota. A total number of 42 samples dataset of Siniperca chuatsi (carnivorous fish) gut microbiota were used to test how the sequencing depth and clustering may affect the alpha and beta diversity patterns of fish intestinal microbiota. Interestingly, we found that the sequencing depth (resampling 1000-11,000 per sample) and the clustering methods (UPARSE and UCLUST) did not bias the estimates of the diversity patterns during the fish development from larva to adult. Although we should acknowledge that a suitable sequencing depth may differ case by case, our finding indicates that a shallow sequencing such as 1000 sequences per sample may be also enough to reflect the general diversity patterns of fish gut microbiota. However, we have shown in the present study that strict pre-processing of the original sequences is required to ensure reliable results. This study provides evidences to help making a strong scientific choice of the sequencing depth and clustering method for future studies on fish gut microbiota patterns, but at the same time reducing as much as possible the costs related to the analysis.
在过去几十年中，16S rRNA基因是用于估计细菌多样性的最常用分子标记之一。然而，测序深度（每个样本从数千到数百万个序列）并不一致，并且用于生成操作分类单元（OTU）的聚类方法在不同研究中也可能不同。这些不一致的前提使得不同研究之间的有效比较变得困难或不可靠。本研究旨在探讨为确保研究鱼类肠道微生物群的稳定多样性模式所需的必要测序深度和聚类方法。总共使用了42个鳜鱼（肉食性鱼类）肠道微生物群样本数据集，以测试测序深度和聚类如何影响鱼类肠道微生物群的α和β多样性模式。有趣的是，我们发现测序深度（每个样本重采样1000 - 11000次）和聚类方法（UPARSE和UCLUST）在鱼类从幼体到成体的发育过程中，不会使多样性模式的估计产生偏差。尽管我们应该承认合适的测序深度可能因情况而异，但我们的发现表明，每个样本1000个序列的浅测序可能也足以反映鱼类肠道微生物群的总体多样性模式。然而，我们在本研究中表明，需要对原始序列进行严格的预处理以确保结果可靠。本研究提供了证据，有助于为未来关于鱼类肠道微生物群模式的研究在测序深度和聚类方法方面做出科学的选择，同时尽可能降低与分析相关的成本。
