运用空间核密度估计和SNP分子分析技术进行结核病发病冷热点地区传播模式探索  

Transmission model of tuberculosis cold and hot spot based on spatial kernel density estimation and SNP molecular analysis

作　　者：崔哲哲[1] 区进[1] 黄莉雯[1] CUI Zhe-zhe;QU Jin;HUANG Li-wen(Guangxi Key Laboratory of Major Infectious Disease Prevention and Control and Biosafety Emergency Response,Tuberculosis Division of Guangxi Center for Disease Control and Prevention,Nanning,Guangxi 530028,China)

机构地区：[1]广西重大传染病防控与生物安全应急响应重点实验,广西壮族自治区疾病预防控制中心结核病防制所,广西南宁530028

出　　处：《现代预防医学》2021年第17期3205-3209,共5页Modern Preventive Medicine

基　　金：国家自然科学基金(81760603);广西自然科学基金(2018GXNSFAA281018);广西医疗卫生适宜技术开发与推广应用项目(S2019067)。

摘　　要：目的通过空间和分子维度的整合分析技术应用,探索不同疫情地区的结核病传播规律和影响因素,为结核病精准防控提供量化依据。方法通过结核病高发地区(热点)和低发地区(冷点)患者现住址坐标点为定位信息的核密度估计,探索空间聚类特征,同时结合菌株的全基因组序列的单个核酸多态性(Single Nucleotide Polymorphisms, SNPs)聚类分析,制作社会网络图。结果基本特征比较发现,热点地区患者较冷点地区的年龄大(rank sum test P=0.002),文化水平低(rank sum test P=0.000),家庭接触来源占比大(χ^(2)=5.122,P=0.024),体重指数小(rank sum test P=0.000),合并糖尿病的比例大(Fisher确切概率法P=0.016);而冷点地区患者外地旅居史占比大(χ^(2)=5.65,P=0.017)。热点和冷点地区侦测到4个空间聚类,主要集中在城乡结合部。热点和冷点地区菌株的全基因组测序共筛选出14 250个SNPs位点,两地区的SNPs矩阵的分子距离中位数分别为897.5和746,两个矩阵的SNP距离差异有统计学意义(rank sum test P=0.000)。热点地区SNPs≤12(近期传播)的分子簇有2个;冷点地区的分子簇有1个。热点地区空间成簇病例42例,分子成簇空间散发病例3例,空间-分子成簇2例;冷点地区有空间成簇26例,分子成簇空间散发病例2例,空间-分子成簇2例。热点和冷点的社会网络密度分别为0.002和0.001;两个地区的空间成簇差异有统计学意义(χ^(2)=3.92,P=0.047)。结论结核病高发热点在本地局域性传播的可能性大于冷点地区,冷点地区存在与外地的基因交流。因此,热点地区的结核病防控应该以本地重点人群主动筛查为主,而冷点地区应加大流动人口排查。Objective To explore the transmission patterns and influence factors of Tuberculosis(TB) in different epidemic areas through the application of integrated analysis technology in spatial and molecular dimensions, and to provide quantitative basis for precise prevention and control of TB. Methods Based on small scale space coordinates of current address information of TB patients in TB high incidence areas(hot spots) and low incidence areas(cold spots), the spatial patterns of infectious TB patients and clustering features were explored, and they combined with a single nucleic acid polymorphisms(SNP) analysis after the whole genome sequencing. Social network diagram had been conducted. Results Through the comparison of basic characteristics, it was found that the patients in hot spots were older(rank sum test P=0.002) than those in cold spots, with lower education level(rank sum test P=0.000), greater family exposure sources(χ^(2)=5.122,P=0.024), smaller body mass index(rank sum test P=0.000) and higher proportion of patients with diabetes(Fisher exact probability test P=0.016). However, the proportion of the patients with recent travel history in cold spots was larger(χ^(2)=5.65,P=0.017). Four spatial clusters were detected, mainly in suburban district. A total of 14 250 SNP sites were screened by whole genome sequencing of strains from hot and cold regions, and the median molecular distance of SNPs matrices from the two regions were 897.5 and 746, respectively. The difference of SNP distance between the two groups of matrices was significant(rank sum test P=0.000). In hot spots, there were two molecular clusters with SNPs ≤12(recently transmission). There was only one molecular cluster in the cold region. There were 42 cases of spatial clustering in hot spots, 3 cases of molecular clustering and 2 cases of spatial-molecular clustering. There were 26 cases of spatial clustering, 2 cases of molecular clustering and 2 cases of spatial-molecular clustering. The density of hot spots and cold spots was 0.002 and 0.

关 键 词：空间 核密度 SNPS 结核病 冷点 热点 

分 类 号：R117[医药卫生—公共卫生与预防医学]