期刊
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA
卷 115, 期 21, 页码 5516-5521出版社
NATL ACAD SCIENCES
DOI: 10.1073/pnas.1802244115
关键词
human genome variation; genomic algorithms; de novo mutations; genomic landscape; coverage map
资金
- Consejo Nacional de Ciencia y Tecnologia [275908]
- US National Science Foundation Award [DBI-1350041]
- US National Institutes of Health Award [R01-HG006677]
- Direct For Biological Sciences
- Div Of Biological Infrastructure [1627442] Funding Source: National Science Foundation
The precise determination of de novo genetic variants has enormous implications across different fields of biology and medicine, particularly personalized medicine. Currently, de novo variations are identified by mapping sample reads from a parent-offspring trio to a reference genome, allowing for a certain degree of differences. While widely used, this approach often introduces false-positive (FP) results due to misaligned reads and mischaracterized sequencing errors. In a previous study, we developed an alternative approach to accurately identify single nucleotide variants (SNVs) using only perfect matches. However, this approach could be applied only to haploid regions of the genome and was computationally intensive. In this study, we present a unique approach, coverage-based single nucleotide variant identification (COBASI), which allows the exploration of the entire genome using second-generation short sequence reads without extensive computing requirements. COBASI identifies SNVs using changes in coverage of exactly matching unique substrings, and is particularly suited for pinpointing de novo SNVs. Unlike other approaches that require population frequencies across hundreds of samples to filter out any methodological biases, COBASI can be applied to detect de novo SNVs within isolated families. We demonstrate this capability through extensive simulation studies and by studying a parent-offspring trio we sequenced using short reads. Experimental validation of all 58 candidate de novo SNVs and a selection of non-de novo SNVs found in the trio confirmed zero FP calls. COBASI is available as open source at https://github.com/Laura-Gomez/COBASI for any researcher to use.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据