FIND:从下一代单端测序数据中快速查找Indel的方法  

FIND:A Fast Method to Detect Indels from Single-end Short Reads

作　　者：宋琳琳[1,2,3,4] 顾朝辉[1,2,3,4] 韦朝春[1,2,5] 

机构地区：[1]上海交通大学生命科学技术学院,上海200240 [2]上海生物信息技术研究中心,上海200235 [3]上海交通大学医学院附属瑞金医院上海血液学研究所,医学基因组学国家重点实验室,上海200025 [4]上海交通大学系统生物医学研究院,上海200240 [5]上海交通大学微生物分子生态学基因组学实验室,上海200240

出　　处：《现代生物医学进展》2010年第1期142-145,共4页Progress in Modern Biomedicine

基　　金：国家863计划(2009AA02Z310);上海市基础重点项目(08JC1416700)

摘　　要：目的:针对下一代测序数据,尤其是单端测序数据,研究快速、准确查找Indel的方法。方法:先与全基因组参考序列进行快速比对,筛选出包含Indel的序列;再对这些序列进行双向的二次比对,确定Indel长度;最后借助长度信息在锁定范围内查找Indel的确切位置和相关信息。结果:本文成功构建FIND(Fast INDel detection system)系统,用于从单端测序数据中查找Indel信息。以模拟测序数据作为测试数据,在12X测试数据情况下,FIND的灵敏度和特异性分别为87.71%和99.66%,而且该性能还随着测序倍数的增加而提升。结论:充分利用比对过程获取的信息,在确定Indle长度的同时也确定出其大致位置,最终在局部范围内实现对单端测序数据中Indle的快速而准确的查找。Objective：To develop a fast and accurate method to detect Indels from thenext-generation sequencing（NGS） data,especially from single-end reads.Methods：In this paper,we First,we mapped reads with a fast mapping system to the human genome reference sequence to get those which may contain Indels.Second,we calculated the length of an Indel by mapping an obtained read again in both directions.Third,we detect the exact positions and relevant information of those Indels in local ranges.Results：We constructed a system called FIND（Fast INdel Detection system） to find indels from single-end short reads quickly.With simulated sequencing data,the sensitivity and specificity of FIND for a 12X sequencing data set were 87.71% and 99.66% respectively.In addition,the performance can be improved when the sequencing coverage is increased.Conclusions：FIND can produce both the length and general position information of potential Indels simmutaniously fast and accurately from single-end sequencing reads.

关 键 词：Indel查找 单端测序 下一代测序 滑动窗口算法 

分 类 号：Q75[生物学—分子生物学] Q78