细菌VI型分泌系统稳定肠道菌群结构的计算模型和方法研究

Development of diseases such as obesity, metabolic syndromes and inflammatory bowel diseases has been tightly associated with the dysbiosis of gut microbiota, which elaborated by many researches. Many species of symbiotic bacteria can secret various effectors via type VI secretion systems(T6SS) to kill other intestinal bacteria or enthetic bacteria, benefiting their colonization and maintaining the structural balance of gut microbiota. However, substantial contributions of T6SS on the stabilization of gut microbiota and functional changes of T6SS genes and effectors in unhealthy gut communities are both keeping unknown. Our team has devoted several years to develop computational methods and tools for bioinformatic applications on bacterial secretion systems. And we also have carried out several studies on the relevance between human gut microbiota and pancreatic cancer or long-term exercises. Based on professional knowledge and basis in bioinformatics, we designed this scientific project to answer above problems. Firstly, in this study, reference datasets of gut metagenomes and T6SS will be constructed by screening public databases, and also by metagenomics sequencing of hundreds of samples collected in our lab. Novel algorithm and tool will be developed to providing fast identification of T6SS in metagenomic datasets. Furthermore, computational methods of stability assessment for gut community will be designed, and then deep comparative analysis will be executed to discover significant differences between healthy and unbalanced gut microbiota. An accurate algorithm to predict the stability of gut community will be developed using feature mining and machine learning techniques. Finally, molecular evolution analysis will be implemented in reference datasets to reveal the influences on the community stability caused by different levels of activation and different directions of evolution of T6SS effectors and immunity proteins. This study is expected to reveal important functional features of T6SS which are crucial to the stability of gut community, and contribute several computational tools which are helpful for exploring the correlation between intestinal bacteria and human diseases.

人体健康受损和疾病的发生发展，与肠道菌群结构和功能失调密切相关。肠道内的共生细菌，可以通过VI型分泌系统（T6SS）分泌效应分子，攻击其它共生或过路细菌，保证自身在肠道定殖，从而在菌群结构平衡中扮演关键角色。T6SS在维持肠道菌群稳定中的具体贡献，以及在与复杂疾病关联的菌群失衡中的功能变化，目前均未得到解释。申请人团队致力于细菌分泌系统的生物信息学研究，并对人体肠道菌群开展前期研究，拟通过本项目来回答上述问题。首先，通过公共数据筛选和宏基因组测序，建立肠道菌群和T6SS参考数据集，开发在宏基因组测序数据中鉴定T6SS的算法和工具；而后，设计肠道菌群稳定性评价方法，通过统计分析和数据挖掘，对健康稳定菌群和结构失衡菌群进行深度比较，阐明二者在T6SS水平上的主要差异，并利用机器学习技术建立肠道菌群稳定性预测算法；最后，通过系统发生分析，揭示T6SS效应分子的激活和进化对肠道菌群稳定性的影响。

本项目建立了细菌VI型分泌系统（T6SS）数据库，通过生物信息学分析，阐明了T6SS效应分子及其免疫蛋白的序列和进化特征，建立了预测T6SS效应分子的机器学习算法，并建立了标准测试数据集，评价了不同预测算法的性能；统计分析了不同来源肠道菌群中T6SS的分布特征，建立了T6SS与肠道菌群稳定水平的关联模型；通过宏基因组测序和生物信息分析，揭示了肠道菌群在爱好运动人群和缺乏运动人群中的差异、以及肠道菌群在不同身体指数人群中，尤其是高BFP人群和低BFP人群中的差异，发现了一些可能作为健康标志物的细菌和代谢物。此外，在此项目的资助下，完成了175万条人类和小鼠染色体外循环DNA（eccDNA）数据的收集、整理和分析，建立了国际上首个eccDNA数据库eccDNAdb，揭示了eccDNA的序列特征，及其在肿瘤组织中的分布特征；另外，在此项目的资助下，收集了超过49万条深圳市孕产数据，通过统计学分析，揭示了深圳这座移民城市中，与孕产妇早产有关的主要危险因子；同时，通过对出生于不同地区的孕产妇的死产数据的分析，揭示了与死产有关的孕妇社会经济学特征，阐明孕产妇社会经济地位的差异是导致早产和死产的重要因素。