微生物组数据分析方法优化及人群肠道菌群分析应用
发布时间：2018-07-18 09:04
【摘要】：研究背景:肠道菌群的分析流程中包含了大量的环节,对最终的分析结果可以起到重要的影响,尤其是在大数据分析方面,诸如不同研究数据如何合并进行分析,以及结果的稳定性和运算的速度等,都是需要验证和优化的问题。另外肠道菌群的人群变异度大,可能会使肠道菌群的研究结果不稳定,一个应对的方案是采用大规模流行病学采样方法研究肠道菌群和宿主关系,从而获得更加可靠和全面的结果,而类似的研究在东方发展中地区还较少。研究目标:本论文首先对微生物组分析中实验操作的影响进行探索,并针对分析流程的稳定性以及运算速度等方面进行优化,并建立完整的分析流程。进一步采用上述方法,分析广东省慢病调查人群肠道菌群,研究肠道菌群与宿主之间的关系,并揭示其代谢综合征特征菌谱。研究方法:第一章我们从四个角度对微生物组数据分析方法进行了研究:1.我们以不同的引物扩增,但使用同一 16SrRNA基因区段为例,观察实验细节对微生物组学分析结果的影响;2.我们以稀释曲线不稳定为切入点,阐述了聚类单元不稳定的原因,并提出较为稳定的聚类方法;3.我们将贪婪重头聚类算法进行了多线程化,试图解决微生物组学大数据分析中聚类速度的问题;4.我们基于现有的一些分析方法和平台,整合了针对微生物组数据分析的流程,并在公共平台上公开。第二章我们在广东省采用聚类抽样法,选择了 14个区/县,每个区/县采用按规模大小成比例的方法(PPS)抽取了 3个街道/镇,在每个街道/镇我们采用PPS方法抽取了两个居委会/村落,并在每个居委会/村落随机抽取了 45户家庭进行调查。我们收集了每位参与者的粪便标本,以及其他的生理指标或社会经济参数。我们采用PERMANOVA法计算了各种宿主信息对肠道菌群变异的解释度,并采用多元线性相关(MaAsLin)方法来计算各类与代谢综合征相关的metadata和具体某些细菌分类之间的关系。研究结果:1.针对微生物组学数据分析方法的验证与优化:实验细节会对微生物组学研究结果造成显著的影响,意味着实验方法严格统一的大规模人群调查对于研究宿主和肠道菌群关系来说是必要的。我们进一步开发了稳定的聚类算法,并将重头贪婪聚类算法进行多线程化来解决计算速度的问题,并基于这些结果建立并公开了一个微生物组学数据分析的流程。2.人群肠道菌群分析:我们在广东省共纳入了 8600位志愿者,并采集了超过100项背景信息。我们发现在广东省地区,肠道菌群与人群的背景信息普遍相关,其中以地理分布的影响最大,而这种地理上的差异可能与当地人食盐习惯相关。其他信息中如年龄,布鲁斯拓评分,体重,尿酸水平,静坐时间,饮食等与肠道菌群变异的相关性也相对较大。在疾病信息中,代谢综合征与肠道菌群的变异相关性最高,其具体的疾病谱特征与发达地区较为相似,但发展中地区的变形菌门细菌显著较高,且和经济发展呈负相关。结论:1.由于微生物组学研究容易受到实验细节的影响,真实实施的流程严格而统一的大规模人群研究是阐述肠菌与宿主关系的重要手段之一,同时我们优化了相关算法,优化了分析方法的稳定性和运算速度,并建立了整合的分析流程。2.肠道菌群与宿主信息普遍相关,其中地理分布是重要的影响因素。我们发现东方发展中地区有其独特的肠道菌群特征,且可能和生活方式协同作用增加代谢性疾病的风险,针对代谢病在快速发展中地区大爆发现象提出了新的解释角度和潜在的干预靶标。
[Abstract]:Research background: the analysis process of intestinal microflora contains a large number of links, which can play an important role in the final analysis, especially in large data analysis, such as how to analyze the combination of different research data, and the stability and speed of the calculation. The large population variation in the population may make the results of the intestinal flora unstable. One solution is to use a large-scale epidemiological sampling method to study the relationship between intestinal flora and host, so as to obtain more reliable and comprehensive results, and similar research is less in the eastern development area. The influence of the experimental operation in the microbiological analysis was explored, and the stability of the analysis process and the speed of operation were optimized, and a complete analysis process was established. Further, the above method was used to analyze the intestinal flora of the Guangdong slow disease survey population, to study the relationship between the intestinal microflora and the host, and to reveal its metabolism. Characteristic Bacteria Spectrum of syndrome. In the first chapter, we studied the data analysis method of microbial group from four angles: 1. we amplified by different primers, but we used the same 16SrRNA gene section as an example to observe the effect of the experimental details on the results of microbiological analysis; 2. we took the dilution curve instability as the breakthrough point. The reasons for the instability of the cluster unit and a more stable clustering method are proposed. 3. we have multithreaded the greedy and heavy head clustering algorithm to solve the problem of clustering speed in the large data analysis of microbiomics. 4., based on some existing analytical methods and platforms, we integrate the flow of data analysis for microbiological groups, and In the public platform, in the second chapter, we adopted cluster sampling in Guangdong Province, selected 14 districts / counties, each district / county took 3 streets / towns by scale and scale method (PPS). In each street / town, we took two neighborhood committees / villages by PPS method, and randomly selected 45 of each neighborhood committee / village. We collected each participant 's fecal specimens, and other physiological or socioeconomic parameters. We used the PERMANOVA method to calculate the interpretation of the diversity of the intestinal flora and the multiple linear correlation (MaAsLin) method to calculate all kinds of metadata related to metabolic syndrome. Specific relationships among certain bacterial classifications. Results: 1. verification and optimization of microbiological data analysis methods: experimental details will have a significant impact on the results of microbiological research, which means that a rigorous and unified mass survey of experimental methods is necessary for the study of the relationship between the host and the intestinal flora. We further developed a stable clustering algorithm, and multithreading the heavy head greedy clustering algorithm to solve the problem of computing speed. Based on these results, we set up and open a microbiome data analysis process for the.2. population analysis of intestinal flora: We included 8600 volunteers in Guangdong Province, and collected more than 10. 0 background information. We found that in Guangdong Province, the intestinal flora is generally related to the background information of the population, and the geographical distribution is the most influential, and the geographical difference may be related to the salt habit of the local people. Other information such as age, Bruce extension, body weight, uric acid level, sitting time, diet and other intestinal flora The correlation of variation is also relatively large. In the disease information, the metabolic syndrome has the highest correlation with the variation of intestinal flora, and the specific characteristics of the disease spectrum are similar to those in the developed areas, but the bacteria of the deformable bacteria in the developing region are significantly higher and have negative correlation with the economic development. Conclusion: 1. because of the microbiological study, the study is easy to be real. It is one of the important means to elaborate the relationship between intestinal bacteria and host, and we optimize the correlation algorithm, optimize the stability and operation speed of the analysis method, and establish an integrated analysis process,.2. intestinal flora and host information. Physical distribution is an important factor. We have found that Eastern developing regions have their unique intestinal microflora characteristics, and they may cooperate with lifestyle to increase the risk of metabolic diseases. New interpretation angles and potential intervention targets are proposed for the rapid development of metabolic diseases in the rapid development of regional outbreak.
【学位授予单位】：南方医科大学
【学位级别】：博士
【学位授予年份】：2017
【分类号】：R371
[Abstract]:Research background: the analysis process of intestinal microflora contains a large number of links, which can play an important role in the final analysis, especially in large data analysis, such as how to analyze the combination of different research data, and the stability and speed of the calculation. The large population variation in the population may make the results of the intestinal flora unstable. One solution is to use a large-scale epidemiological sampling method to study the relationship between intestinal flora and host, so as to obtain more reliable and comprehensive results, and similar research is less in the eastern development area. The influence of the experimental operation in the microbiological analysis was explored, and the stability of the analysis process and the speed of operation were optimized, and a complete analysis process was established. Further, the above method was used to analyze the intestinal flora of the Guangdong slow disease survey population, to study the relationship between the intestinal microflora and the host, and to reveal its metabolism. Characteristic Bacteria Spectrum of syndrome. In the first chapter, we studied the data analysis method of microbial group from four angles: 1. we amplified by different primers, but we used the same 16SrRNA gene section as an example to observe the effect of the experimental details on the results of microbiological analysis; 2. we took the dilution curve instability as the breakthrough point. The reasons for the instability of the cluster unit and a more stable clustering method are proposed. 3. we have multithreaded the greedy and heavy head clustering algorithm to solve the problem of clustering speed in the large data analysis of microbiomics. 4., based on some existing analytical methods and platforms, we integrate the flow of data analysis for microbiological groups, and In the public platform, in the second chapter, we adopted cluster sampling in Guangdong Province, selected 14 districts / counties, each district / county took 3 streets / towns by scale and scale method (PPS). In each street / town, we took two neighborhood committees / villages by PPS method, and randomly selected 45 of each neighborhood committee / village. We collected each participant 's fecal specimens, and other physiological or socioeconomic parameters. We used the PERMANOVA method to calculate the interpretation of the diversity of the intestinal flora and the multiple linear correlation (MaAsLin) method to calculate all kinds of metadata related to metabolic syndrome. Specific relationships among certain bacterial classifications. Results: 1. verification and optimization of microbiological data analysis methods: experimental details will have a significant impact on the results of microbiological research, which means that a rigorous and unified mass survey of experimental methods is necessary for the study of the relationship between the host and the intestinal flora. We further developed a stable clustering algorithm, and multithreading the heavy head greedy clustering algorithm to solve the problem of computing speed. Based on these results, we set up and open a microbiome data analysis process for the.2. population analysis of intestinal flora: We included 8600 volunteers in Guangdong Province, and collected more than 10. 0 background information. We found that in Guangdong Province, the intestinal flora is generally related to the background information of the population, and the geographical distribution is the most influential, and the geographical difference may be related to the salt habit of the local people. Other information such as age, Bruce extension, body weight, uric acid level, sitting time, diet and other intestinal flora The correlation of variation is also relatively large. In the disease information, the metabolic syndrome has the highest correlation with the variation of intestinal flora, and the specific characteristics of the disease spectrum are similar to those in the developed areas, but the bacteria of the deformable bacteria in the developing region are significantly higher and have negative correlation with the economic development. Conclusion: 1. because of the microbiological study, the study is easy to be real. It is one of the important means to elaborate the relationship between intestinal bacteria and host, and we optimize the correlation algorithm, optimize the stability and operation speed of the analysis method, and establish an integrated analysis process,.2. intestinal flora and host information. Physical distribution is an important factor. We have found that Eastern developing regions have their unique intestinal microflora characteristics, and they may cooperate with lifestyle to increase the risk of metabolic diseases. New interpretation angles and potential intervention targets are proposed for the rapid development of metabolic diseases in the rapid development of regional outbreak.
【学位授予单位】：南方医科大学
【学位级别】：博士
【学位授予年份】：2017
【分类号】：R371
【相似文献】
相关期刊论文 前10条
1 缪其宏;新兵在服役适应期的肠道菌群[J];解放军预防医学杂志;1992年01期
2 刘华;黄湘宁;许莉;;成都市中老年人肠道菌群比例的初步探讨[J];现代预防医学;2007年20期
3 王新文;阿基业;曹蓓;刘林生;赵春艳;陈晓虎;王广基;;肠道菌群对某些疾病及药物疗效与毒性的影响[J];药学进展;2012年05期
4 张晓婧;曾本华;刘智伟;廖振林;方祥;魏泓;;两种不同品系小鼠的人源菌群模型的建立与肠道菌群的比较[J];中国微生态学杂志;2013年04期
5 陈秀琴;黄小洁;石达友;郭世宁;;中药与肠道菌群相互作用的研究进展[J];中草药;2014年07期
6 ;肠道菌群 你所不知的那部分“自己”[J];科学之友(上旬);2013年06期
7 阿梅;漫话肠道菌群平衡[J];中老年保健;1997年02期
8 尹军霞,林德荣;肠道菌群与疾病[J];生物学通报;2004年03期
9 王建阳;关注住院病人的肠道菌群[J];现代检验医学杂志;2005年04期
10 ;欧盟启动肠道菌群基因组研究工程[J];中国食品学报;2008年03期
相关会议论文 前10条
1 曹虹;郝小燕;彭亮;方幸幸;;肠道菌群与机体代谢及相关疾病——感染微生态学的研究进展[A];新发和再发传染病防治热点研讨会论文集[C];2011年
2 王保红;李e，
本文编号：2131419
本文链接：https://www.wllwen.com/shoufeilunwen/yxlbs/2131419.html
教材专著
