Veterinary Microbiology and Epidemiology, Department of Veterinary Biosciences, Faculty of Veterinary Medicine, University of Helsinki, Helsinki, Finland.
J Clin Gastroenterol. 2010 Sep;44 Suppl 1:S2-5. doi: 10.1097/MCG.0b013e3181e5018f.
We describe advanced approaches for the computational meta-analysis of a collection of independent studies, including over 1000 phylogenetic array datasets, as a means to characterize the variability of human intestinal microbiota.
The human intestinal microbiota is a complex microbial community, consisting of several thousands of phylotypes, is specific for each individual, and impacts health and disease. We have developed a phylogenetic microarray, the Human Intestinal Tract Chip, to address the microbial diversity of the intestinal microbiota and used this tool to generate large datasets. It is of significant interest to use these datasets to be able to provide relations between microbial taxa, describe the extent and type of variability of the microbiota in the human gut, and establish relations between microbial taxa and their interaction with the host, intestinal location, or genotype.
We present an advanced computational meta-analysis approach for studying human intestinal microbiota, outline the advantages and disadvantages of such a meta-analysis, and reflect it to analogous approaches in other fields. Finally, we illustrate the potential of this meta-analysis by identifying salient signatures of site-specific microbial communities, describe impact of genotype, and provide first examples of relevant relations between microbial taxa.
We are in the process of designing and applying appropriate methods for carrying out a full meta-analysis of the present data. Beyond that, the next large challenges in future meta-analyses lie in the integration of data from several heterogeneous measurement methods such as next generation sequencing techniques, metaproteomics, or metabolomics.
We have shown the feasibility of an advanced computational meta-analysis of the large datasets derived from the human intestinal microbiota.
我们描述了一种对独立研究集合进行计算元分析的高级方法，包括超过 1000 个基于基因体的微阵列数据集，以用于描述人类肠道微生物组的可变性。
人类肠道微生物组是一种复杂的微生物群落，由数千个生物型组成，具有个体特异性，并影响健康和疾病。我们开发了一种基于基因体的微阵列，即人类肠道芯片，用于解决肠道微生物组的微生物多样性问题，并使用该工具生成了大量数据集。利用这些数据集来提供微生物类群之间的关系，描述人类肠道微生物组的可变性程度和类型，以及建立微生物类群与宿主、肠道位置或基因型之间的相互关系，具有重要意义。
我们提出了一种用于研究人类肠道微生物组的高级计算元分析方法，概述了这种元分析的优缺点，并将其与其他领域的类似方法进行了比较。最后，我们通过识别特定部位微生物群落的显著特征、描述基因型的影响以及提供微生物类群之间相关关系的初步示例，说明了这种元分析的潜力。
我们正在设计和应用适当的方法，对目前的数据进行全面的元分析。除此之外，未来元分析的下一个大挑战在于整合来自多种异构测量方法的数据，如下一代测序技术、宏蛋白质组学或代谢组学。
我们已经证明了对人类肠道微生物组的大型数据集进行高级计算元分析的可行性。
