一种基于动态框长的局部祖先推断模型研究  

A research of local ancestry inference model based on dynamic segment length

作　　者：吴杰[1,2] WU Jie(Institute of Chinese Materia Medica,China Academy of Chinese Medicine Sciences,Beijing 100700,China;College of Biological Sciences,China Agricultural University,Beijing 100193,China)

机构地区：[1]中国中医科学院中药研究所,北京100700 [2]中国农业大学生物学院,北京100193

出　　处：《生物信息学》2024年第4期262-269,共8页Chinese Journal of Bioinformatics

基　　金：中国中医科学院优秀青年科技人才培养专项(No.ZZ15-YQ-036)

摘　　要：由于人群混合和遗传重组,基因组中每条染色体可以看作来自不同遗传背景的祖先染色体的嵌合。局部祖源推断(Local ancestry inference,LAI)旨在利用遗传标记信息推断基因组中每条染色体片段的祖先来源,是遗传分析的重要内容。为了简化LAI的设计过程,现有的方法大多使用固定大小的窗口,忽略了待测个体基因组中真实的重组断点。本研究设计了一种基于k-mers的策略寻找重组断点,在划分每个单倍体中不定长的重组窗口的基础上结合隐马尔科夫链(Hidden markov models,HMMs)进行局部祖源推断。模拟显示本研究模型WAdmix的准确率近似于包括RFMix在内的主流工具。Due to population mixing and genetic recombination,each chromosome in the genome can be regarded as a chimera of ancestral chromosomes from different genetic backgrounds.Local Ancestry Inference(LAI)is an important part of genetic analysis,which infers the ancestral origin of each chromosome segment in the genome by using genetic marker information.In order to simplify the process of LAI design,most of the existing methods use fixed⁃size windows and ignore the real recombination breakpoints in the individual genome to be tested.In this study,we design a strategy based on k⁃mers to find recombination breakpoints,and combine Hidden Markov models(HMMs)to infer local ancestry based on splitting of variable⁃length recombination windows in each haploid.Simulation results show that the accuracy of Wadmix is similar to that of other mainstream tools including RFMix.

关 键 词：HMMs k-mers 局部祖先推断 不定长片段 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]