Format

Send to

Choose Destination
PLoS One. 2013 Apr 23;8(4):e61292. doi: 10.1371/journal.pone.0061292. Print 2013.

On the power and the systematic biases of the detection of chromosomal inversions by paired-end genome sequencing.

Author information

1
Institut de Biotecnologia i de Biomedicina, Universitat Autònoma de Barcelona, Bellaterra, Barcelona, Spain.

Abstract

One of the most used techniques to study structural variation at a genome level is paired-end mapping (PEM). PEM has the advantage of being able to detect balanced events, such as inversions and translocations. However, inversions are still quite difficult to predict reliably, especially from high-throughput sequencing data. We simulated realistic PEM experiments with different combinations of read and library fragment lengths, including sequencing errors and meaningful base-qualities, to quantify and track down the origin of false positives and negatives along sequencing, mapping, and downstream analysis. We show that PEM is very appropriate to detect a wide range of inversions, even with low coverage data. However, ≥% of inversions located between segmental duplications are expected to go undetected by the most common sequencing strategies. In general, longer DNA libraries improve the detectability of inversions far better than increments of the coverage depth or the read length. Finally, we review the performance of three algorithms to detect inversions--SVDetect, GRIAL, and VariationHunter--, identify common pitfalls, and reveal important differences in their breakpoint precisions. These results stress the importance of the sequencing strategy for the detection of structural variants, especially inversions, and offer guidelines for the design of future genome sequencing projects.

PMID:
23637806
PMCID:
PMC3634047
DOI:
10.1371/journal.pone.0061292
[Indexed for MEDLINE]
Free PMC Article

Supplemental Content

Full text links

Icon for Public Library of Science Icon for PubMed Central
Loading ...
Support Center