Discovery of large genomic inversions using long range information

Date
2017
Authors
Rasekh, M. E.
Chiatante, G.
Miroballo, M.
Tang, J.
Ventura M.
Amemiya, C. T.
Eichler, E. E.
Antonacci, F.
Alkan C.
Editor(s)
Advisor
Supervisor
Co-Advisor
Co-Supervisor
Instructor
Source Title
BMC Genomics
Print ISSN
Electronic ISSN
1471-2164
Publisher
BioMed Central Ltd.
Volume
18
Issue
65
Pages
1 - 12
Language
English
Journal Title
Journal ISSN
Volume Title
Series
Abstract

Although many algorithms are now available that aim to characterize different classes of structural variation, discovery of balanced rearrangements such as inversions remains an open problem. This is mainly due to the fact that breakpoints of such events typically lie within segmental duplications or common repeats, which reduces the mappability of short reads. The algorithms developed within the 1000 Genomes Project to identify inversions are limited to relatively short inversions, and there are currently no available algorithms to discover large inversions using high throughput sequencing technologies. Results: Here we propose a novel algorithm, Valor, to discover large inversions using new sequencing methods that provide long range information such as 10X Genomics linked-read sequencing, pooled clone sequencing, or other similar technologies that we commonly refer to as long range sequencing. We demonstrate the utility of Valor using both pooled clone sequencing and 10X Genomics linked-read sequencing generated from the genome of an individual from the HapMap project (NA12878). We also provide a comprehensive comparison of Valor against several state-of-the-art structural variation discovery algorithms that use whole genome shotgun sequencing data. Conclusions: In this paper, we show that Valor is able to accurately discover all previously identified and experimentally validated large inversions in the same genome with a low false discovery rate. Using Valor, we also predicted a novel inversion, which we validated using fluorescent in situ hybridization. Valor is available at https://github.com/BilkentCompGen/Valor. © 2017 The Author(s).

Course
Other identifiers
Book Title
Citation
Published Version (Please cite this version)