CLOVE: classification of genomic fusions into structural variation events
Details
Publication Year 2017-07-20,Volume 18,Issue #1,Page 346
Journal Title
BMC Bioinformatics
Publication Type
Journal Article
Abstract
BACKGROUND: A precise understanding of structural variants (SVs) in DNA is important in the study of cancer and population diversity. Many methods have been designed to identify SVs from DNA sequencing data. However, the problem remains challenging because existing approaches suffer from low sensitivity, precision, and positional accuracy. Furthermore, many existing tools only identify breakpoints, and so not collect related breakpoints and classify them as a particular type of SV. Due to the rapidly increasing usage of high throughput sequencing technologies in this area, there is an urgent need for algorithms that can accurately classify complex genomic rearrangements (involving more than one breakpoint or fusion). RESULTS: We present CLOVE, an algorithm for integrating the results of multiple breakpoint or SV callers and classifying the results as a particular SV. CLOVE is based on a graph data structure that is created from the breakpoint information. The algorithm looks for patterns in the graph that are characteristic of more complex rearrangement types. CLOVE is able to integrate the results of multiple callers, producing a consensus call. CONCLUSIONS: We demonstrate using simulated and real data that re-classified SV calls produced by CLOVE improve on the raw call set of existing SV algorithms, particularly in terms of accuracy. CLOVE is freely available from http://www.github.com/PapenfussLab .
Publisher
BioMed Central
Research Division(s)
Bioinformatics
PubMed ID
28728542
NHMRC Grants
NHMRC/1054618NHMRC/1116955
Terms of Use/Rights Notice
Refer to copyright notice on published article.


Creation Date: 2017-08-30 02:22:38
Last Modified: 2017-09-04 03:27:56
An error has occurred. This application may no longer respond until reloaded. Reload 🗙