Open Access Highly Accessed Research

Haplotype-assisted accurate non-invasive fetal whole genome recovery through maternal plasma sequencing

Shengpei Chen12, Huijuan Ge1, Xuebin Wang1, Xiaoyu Pan13, Xiaotian Yao1, Xuchao Li1, Chunlei Zhang1, Fang Chen1, Fuman Jiang1, Peipei Li1, Hui Jiang1, Hancheng Zheng1, Lei Zhang1, Lijian Zhao1, Wei Wang1, Songgang Li1, Jun Wang1, Jian Wang1, Huanming Yang1, Yingrui Li1* and Xiuqing Zhang1*

Author Affiliations

1 BGI-Shenzhen, Shenzhen 518083, China

2 State Key Laboratory of Bioelectronics, School of Biological Science and Medical Engineering, Southeast University, Nanjing 210096, China

3 School of Bioscience and Bioengineering, South China University of Technology, Guangzhou 510000, China

For all author emails, please log on.

Genome Medicine 2013, 5:18  doi:10.1186/gm422


See related Commentary article: http://www.biomedcentral.com/1741-7015/11/56

Published: 27 February 2013

Abstract

Background

The applications of massively parallel sequencing technology to fetal cell-free DNA (cff-DNA) have brought new insight to non-invasive prenatal diagnosis. However, most previous research based on maternal plasma sequencing has been restricted to fetal aneuploidies. To detect specific parentally inherited mutations, invasive approaches to obtain fetal DNA are the current standard in the clinic because of the experimental complexity and resource consumption of previously reported non-invasive approaches.

Methods

Here, we present a simple and effective non-invasive method for accurate fetal genome recovery-assisted with parental haplotypes. The parental haplotype were firstly inferred using a combination strategy of trio and unrelated individuals. Assisted with the parental haplotype, we then employed a hidden Markov model to non-invasively recover the fetal genome through maternal plasma sequencing.

Results

Using a sequence depth of approximately 44X against a an approximate 5.69% cff-DNA concentration, we non-invasively inferred fetal genotype and haplotype under different situations of parental heterozygosity. Our data show that 98.57%, 95.37%, and 98.45% of paternal autosome alleles, maternal autosome alleles, and maternal chromosome X in the fetal haplotypes, respectively, were recovered accurately. Additionally, we obtained efficient coverage or strong linkage of 96.65% of reported Mendelian-disorder genes and 98.90% of complex disease-associated markers.

Conclusions

Our method provides a useful strategy for non-invasive whole fetal genome recovery.