基于高通量测序数据的插入/删除新突变检测方法

De Novo Indel detection method based on high-throughput sequencing data

插入/删除新突变是一种重要的新突变形式,与多种人类疾病的发生密切相关。随着高通量测序技术的迅猛发展,基于高通量测序数据进行插入/删除新突变检测已成为常规手段,但由于测序错误以及reads比对错误的影响,已有的检测方法通常存在错误率较高的问题。本文提出一种基于Adaboost的插入/删除新突变检测方法,旨在对常用的新突变检测方法产生的插入/删除新突变检测结果进行过滤,在确保基本不损失敏感度的前提下,显著降低错误发现率。

De Novo Indel is an important form of de novo mutation,and it is closely related to the occurrence of a variety of human diseases.With the development of high-throughput sequencing technology,using high-throughput data to detect De Novo Indels has become routine.However,due to the sequencing errors and the reads alighment errors,existing detection methods usually yield high error.This paper proposes a new De Novo Indel detection method based on Adaboost.This method is designed to filter De Novo Indels detected by common De Novo Indel detection methods,which can significantly reduce the false discovery rate without sacrificing the sensitivity.