Motivation: Identification of structural variants (SVs) in sequence data results in a large number of false positive calls using existing software, which overburdens subsequent validation.
Results: Simulations using RAPTR-SV and other, similar algorithms for SV detection revealed that RAPTR-SV had superior sensitivity and precision, as it recovered 66.4% of simulated tandem duplications with a precision of 99.2%. When compared with calls made by Delly and LUMPY on available datasets from the 1000 genomes project, RAPTR-SV showed superior sensitivity for tandem duplications, as it identified 2-fold more duplications than Delly, while making ∼85% fewer duplication predictions.
Availability and implementation: RAPTR-SV is written in Java and uses new features in the collections framework in the latest release of the Java version 8 language specifications. A compiled version of the software, instructions for usage and test results files are available on the GitHub repository page: https://github.com/njdbickhart/RAPTR-SV.
Contact: derek.bickhart@ars.usda.gov.
Published by Oxford University Press 2015. This work is written by US Government employees and is in the public domain in the US.