鼠疫自然疫源地局部生态与鼠疫菌基因组变异的关联分析

发布时间：2018-04-25 06:19

本文选题：生态流行病学 + 基因组流行病学　；参考：《中国人民解放军军事医学科学院》2015年博士论文

【摘要】：研究目的近十年,由于测序技术的迅速发展,使得我们能够在短时间内低成本完成大量基因组的测序,也为细菌的群体遗传学研究和表型性状与遗传基础的关联分析研究提供了前所未有的机遇。与此同时,如何合理处理和分析海量基因组数据,给生物统计学带来了新的机遇和挑战,也极大的促进了该学科的发展。本研究以鼠疫菌为研究对象,将生态流行病学涉及的气候与鼠疫疫情的研究与基因组流行病学涉及的鼠疫菌全基因组范围的遗传变异综合在一起,探讨鼠疫菌在局部地区气候影响下的进化规律。这样的跨学科研究,开启了鼠疫菌研究的新领域,为其它人畜共患病原菌研究提供参考。研究对象鼠疫菌在人类历史上曾有三次大流行,造成上亿人死亡。目前虽然没有在人类中大规模流行,但仍存在于鼠疫自然疫源地中,并可通过蚤等媒介在啮齿类动物间传播流行。乌苏地区位于新疆维吾尔自治区,属于北天山灰旱獭、长尾黄鼠鼠疫疫源地,是重点的疫情监测地,在该地区自然环境中动物间鼠疫常年流行。本研究从该地区保存的120株历史菌株中筛选了102株菌进行测序,并最终确定93株菌用于本研究分析。同时,为了分析在50年的历史中环境因素和宿主媒介动态变化,本研究还收集和使了用该地区的鼠疫监测信息和气候环境数据。研究内容与结果-局部地区鼠疫菌进化对采样得到的乌苏地区93株鼠疫菌鉴定出166个可靠SNP位点。通过这些位点,可以对这些菌株构建系统发育树,分析其进化关系。系统发育分析表明,乌苏地区的鼠疫菌可以分为2个大群,分别位于古尔图地区的两片区域(记为A和B)(共78株)和巴音沟地区(共14株),另外还有1株距离两个群关系都比较远,可能是由于两个地区地理隔离导致不同种群的形成。另外,本研究还将93株菌与全球鼠疫菌系统发育树做比较,发现这93株菌均可定位在0.ANT1分支中。考虑到古尔图和巴音沟地区距离较远,海拔相差上千米,生态环境差别较大,本研究将焦点聚集在古尔图地区分离的78株鼠疫菌以及该地区的鼠疫生境中。古尔图地区鼠疫分离株中共鉴定出54个SNP。通过多种不同建树方法,均得到一致系统发育关系。78株菌可以分为3个主要群,并可进一步细分为Group 1.1,1.2,2.1,2.2和3。通过观察各群在时间上的动态变化,我们发现在A地区Group 1明显被Group 2所逐渐替代,发生时间约在1987年到1989年之间,1990年后Group1在A地区仅观测到1例。通过BEAST2推测各种群分化时间,Group 1和Group 2两群在1929年(95%置信区间为1900到1954年)产生分化。种群的替代可能是由于某种外部因素影响,导致另一个种群Group 2取得优势地位。通过有效种群大小和有效种群复制率分析,可以发现在1983年到1989年中,有效复制率明显降低,而有效种群大小也在1987年后明显下降,这些事实都表明在这个时期内可能有异常的外部环境变化,形成选择压力,从而改变了鼠疫菌的种群构成。基于鼠疫菌的SNP、Indel和附加基因组变异分析结果,本研究分析了基因组中受自然选择的区域,最终发现12个变异热区,并对其进行了详细注释。其中受选择最显著的热区位于rpo Z编码基因,基因总长只有276bp,发生了8个变异,其变异发生率(8/276≈2.90×10-2)远高于78株菌的全基因组变异发生率(128/4653728≈2.73×10-5)。rpo Z基因与鼠疫菌的生长速率以及在蚤体内生物膜形成有关,因而这些变异可能会影响鼠疫菌在跳蚤和老鼠中间的传播,但由于生长速率降低这些变异均未在种群中固定下来。-局部地区鼠疫疫情与气候环境变化关联为了充分挖掘疫情监测数据,本研究对各指标首先进行了两两之间的相关分析。对蚤指数、鼠密度和血清阳性率进行Pearson相关分析,结果显示三者之间相关性不高,甚至相关系数都没有超过0.5。理论上三者之间应该是有着直接的影响作用,因此相关度不高可能是由于其它因素如气候和环境导致。三个指标的自相关和两两之间的互相关也不存在规律性,说明这三个指标间在一定时延下也无显著的相关性。参考与古尔图地区相临的哈萨克斯坦地区鼠疫疫情规律所做的监测数据随时间变化图表示,该地区不存在与哈萨克斯坦地区相同的疫情规律,仍需要进一步深入研究其成因。对气候环境数据的探索性分析中,首先分析了降水、温度和NDVI数据的季节性,通过谱分析确定了有且仅有以1年为单位的周期分量。通过对时间序列季节项分解过滤周期分量后,得到了不包含周期波动的三项气候环境数据。这三个时间序列(季节分量调整后三项气候环境数据)本身前后数据之间仍存在一定的相关性,为了能有效鉴定在50年间何时出现异常气候情况,则需要将该规律部分进一步分解。通过使用时间序列中ARIMA模型方法,对这三项数据进一步分解后,得到了接近高斯白噪声,前后之间相关性也不存在。这里剩余的残差项就是气候环境数据在排除规律性因素后剩余的随机分量,此数据中的异常数据即为气候环境异常点。通过广义极端学生化偏差(ESD)方法,最终鉴定出8个温度异常值,9个降水异常值以及1个NDVI异常值。异常点的分布并不均匀,在1986年到1990年间,多次出现降水异常以及温度异常。为研究监测指标之间以及与气候环境指标之间的相互关系,希望通过建立定量化的回归方程来研究这些指标之间的规律。由于涉及变量数目过多,因而采用遗传算法解决变量筛选问题,以AICc为标准选择最优模型。通过该方法得到的方程揭示了监测指标与当年和前一年的监测指标以及气候环境指标中影响最大的分量。所建立的三个方程中,都包含有鼠密度指标,说明鼠密度在鼠疫菌流行的生态系统中发挥着重要的作用。将气候环境异常与基因组变异相联系,发现1986年到1990年间,是气候异常频发的时段,正好对应了检测到的鼠疫菌变异最多的一段时间,可能是导致鼠疫菌种群克隆群转换的原因。结论与意义本研究通过对局部地区——古尔图地区鼠疫菌的基因组进化、动物鼠疫疫情与气候变化的相关和回归分析,给出了鼠疫菌所处生态系统在历史中的动态变化过程,为后续深入探索鼠疫菌在鼠疫自然疫源地的流行规律做出铺垫。rpo Z基因很可能与鼠疫菌环境适应性密切相关,可能会导致鼠疫菌的流行强度增加。结合气候与环境数据,发现连续的气候异常同鼠疫菌克隆群转换时间恰好一致,可能是由于极端天气导致鼠疫菌种群下降,从而引起鼠疫菌种群克隆群变化。这提示我们应当注意连续极端天气下鼠疫菌发生的变异,预防具有高传染性鼠疫菌种群出现。通过回归分析,发现宿主密度在整个鼠疫菌流行的生态系统中,可能是最重要的一项因素。本研究所得到的线性方程,可用于对古尔图地区鼠疫疫情的流行进行预测。
[Abstract]:For the last ten years, the rapid development of sequencing technology has made it possible for us to complete a large number of genome sequencing in a short period of time. It also provides unprecedented opportunities for bacterial population genetics research and association analysis of phenotypic traits and genetic bases. At the same time, how to deal with and analyze mass bases reasonably The data of the group have brought new opportunities and challenges to biometrics, which also greatly promoted the development of the subject. This study takes the Yersinia pestis as the object of study, and combines the study of the epidemic of ecologic epidemiology with the epidemic of plague and the genetic variation of the scope of the whole group of Yersinia pestis involved in the genome epidemiology. The evolution of Phytophthora under the influence of local climate. This interdisciplinary study opens a new field for the study of Yersinia pestis and provides reference for the study of other zoonosis. The research object of Yersinia pestis has three pandemics in human history, causing hundreds of millions of deaths. It exists in the natural foci of plague and can spread among rodents through fleas and other vectors. The Wusu area is located in the Xinjiang Uygur Autonomous Region, belongs to the Northern Tianshan gray marmot, the plague foci of the long tail yellow squirrel plague, the key epidemic monitoring site, and the perennial epidemic of the plague among the animals in the natural environment of the region. This study is preserved from this area. Of the 120 strains of historical strains, 102 strains were sequenced and 93 strains were determined for this study. In order to analyze the dynamic changes of environmental factors and host media in the history of 50 years, this study also collected and made the monitoring information of plague and climate and environment data in the area. Phytophthora evolution has identified 166 reliable SNP loci for 93 strains of Yersinia pestis in Wusu area. Through these loci, we can construct phylogenetic trees and analyze their evolutionary relationships. Phylogenetic analysis shows that the Yersinia pestis can be divided into 2 large groups in the two regions of the guertu region (A and B). A total of 78 strains and 14 strains of Ba Yin gully (a total of 1 strains) are far from two groups, which may be caused by geographical isolation in two regions. In addition, the study also compares 93 strains with the global Yersinia phylogenetic tree, and found that these 93 strains can be located in the 0.ANT1 branch. The area of the bayergou area is far distance, the altitude difference is thousands of meters, and the ecological environment is different. This study focuses on 78 strains of Yersinia pestis isolated in guertu area and the plague habitat in this area. The isolated strains of plague in guertu area have identified 54 SNP. through a variety of different methods of tree building, all of which are consistent with the phylogenetic relationship of.78 strain. Bacteria can be divided into 3 main groups and can be further subdivided into Group 1.1,1.2,2.1,2.2 and 3. by observing the dynamic changes in time. We found that Group 1 in A area was gradually replaced by Group 2. The occurrence time was between 1987 and 1989. After 1990, only 1 cases were observed in A region. All kinds of groups were conjectured through BEAST2. Differentiation time, Group 1 and Group 2 two groups were differentiated in 1929 (95% confidence interval from 1900 to 1954). The substitution of the population may be due to some external factors, resulting in the dominant position of another population Group 2. The effective population size and the effective population replication rate can be found to be effectively replicated from 1983 to 1989. The rate of the population decreased significantly, and the effective population size decreased significantly after 1987. These facts indicate that there may be abnormal external environment changes during this period, forming selection pressure, thus changing the population composition of Yersinia pestis. Based on the analysis of SNP, Indel and additional gene group variation of Yersinia pestis, this study analyzed the genome in the genome. In the area of natural selection, 12 variation heat regions were found and the most significant heat areas were selected as the RPO Z coding gene, and the total gene length was only 276bp, and 8 variations occurred. The mutation rate (8/276 2.90 * 10-2) was far higher than that of 78 strains (128/4653728 2.73 * 10-5).Rpo. The Z gene is related to the growth rate of Yersinia pestis and the formation of biofilm in the flea body, so these variations may affect the spread of Yersinia pestis between fleas and mice, but these variations are not fixed in the population due to the decrease of the growth rate. - the epidemic of plague in local areas is associated with the climate change in order to fully excavate the epidemic situation. Monitoring data, this study first carried out 22 correlation analysis of each index. The Pearson correlation analysis was carried out on the flea index, rat density and the positive rate of serum. The results showed that the correlation between the three was not high, even the correlation coefficient did not exceed the theory of the 0.5., which should be directly affected by the three, so the correlation was not high. It may be due to other factors such as climate and environment. The autocorrelation of the three indicators and the intercorrelation between the 22 are also not regular, indicating that there is no significant correlation between the three indicators under a certain delay. The map indicates that the region does not have the same epidemic situation as the Kazakhstan area and still needs to further study its causes. In the exploratory analysis of climate and environmental data, the seasonal characteristics of precipitation, temperature and NDVI data are analyzed, and the periodic components with only 1 years as units are determined by spectral analysis. After the sequence seasonal term is decomposed and filtered, the three climatic environment data, which does not contain periodic fluctuations, are obtained. The three time series (three climatic environment data after the seasonal component adjustment) still have a certain correlation between the data and the data itself. In order to effectively identify when the abnormal climate occurs in the period of 50, it needs to be taken. The rule part is further decomposed. By using the ARIMA model method in the time series, after the further decomposition of the three data, we get close to Gauss white noise, and there is no correlation between before and after. The remaining residual term is the residual random component after the climatic environment data are excluded from the regular factors and the abnormal data in this data By means of generalized extreme student deviation (ESD) method, 8 temperature anomaly values, 9 anomalous values of precipitation and 1 NDVI anomalies are identified. The distribution of abnormal points is not uniform. From 1986 to 1990, there are many abnormal precipitation and temperature anomalies. The relationship between indexes is expected to be studied by establishing a quantitative regression equation. As the number of variables is too large, the genetic algorithm is used to solve the problem of variable selection, and the optimal model is selected as the standard of AICc. The formula obtained through this method reveals the monitoring index and the year and the previous year. The three equations which have the greatest impact on the monitoring indicators and climate and environmental indicators all contain mouse density indicators, indicating that the rat density plays an important role in the ecological system of Yersinia pestis. It is found that between 1986 and 1990, the climate and environmental anomalies are associated with genomic variation, and it is a period of abnormal climate. It is exactly corresponding to the time of the most mutation of the detected Yersinia pestis, which may be the cause of the transformation of the clone group of the Yersinia pestis population. The dynamic change process of the ecosystem in the history, in order to further explore the epidemic law of Yersinia pestis in the plague natural foci, the.Rpo Z gene may be closely related to the environmental adaptability of the Yersinia pestis, which may lead to the increase of the epidemic intensity of the Yersinia pestis. The conversion time of bacteria clones is exactly the same. It may be caused by the decline of Yersinia pestis population due to extreme weather, which leads to the change of the population clones of Yersinia pestis. This suggests that we should pay attention to the variation of Yersinia pestis in the continuous extreme weather and prevent the emergence of highly contagious plague strains. The epidemic of Yersinia may be the most important factor in the epidemic. The linear equation obtained in this study can be used to predict the epidemic of plague in the guertu region.

【学位授予单位】：中国人民解放军军事医学科学院
【学位级别】：博士
【学位授予年份】：2015
【分类号】：R516.8

【相似文献】