PeSV-Fisher: identification of somatic and non-somatic structural variants using nextgeneration sequencing data

Escaramís, Geòrgia; Tornador Antolin, Cristian, 1979-; Bassaganyas Bars, Laia, 1985-; Rabionet Janssen, Raquel; Tubio, José M. C.; Martínez Fundichely, Alexander, 1978-; Cáceres Aguilar, Mario; Gut, Marta; Ossowski, Stephan; Estivill, Xavier, 1955-

doi:http://dx.doi.org/10.1371/journal.pone.0063377

PeSV-Fisher: identification of somatic and non-somatic structural variants using nextgeneration sequencing data

License

Citation

Escaramís G, Tornador C, Bassaganyas L, Rabionet R, Tubio JM,/nMartínez-Fundichely A et al. PeSV-Fisher: identification of somatic and non-somatic structural variants using next generation sequencing data. PLoS One. 2013 May 21;8(5):e63377. DOI: 10.1371/journal.pone.0063377

Abstract

Next-generation sequencing technologies expedited research to develop efficient computational tools for the identification of structural variants (SVs) and their use to study human diseases. As deeper data is obtained, the existence of higher complexity SVs in some genomes becomes more evident, but the detection and definition of most of these complex rearrangements is still in its infancy. The full characterization of SVs is a key aspect for discovering their biological implications. Here we present a pipeline (PeSV-Fisher) for the detection of deletions, gains, intra- and inter-chromosomal translocations, and inversions, at very reasonable computational costs. We further provide comprehensive information on co-localization of SVs in the genome, a crucial aspect for studying their biological consequences. The algorithm uses a combination of methods based on paired-reads and read-depth strategies. PeSV-Fisher has been designed with the aim to facilitate identification of somatic variation, and, as such, it is capable of analysing two or more samples simultaneously, producing a list of non-shared variants between samples. We tested PeSV-Fisher on available sequencing data, and compared its behaviour to that of frequently deployed tools (BreakDancer and VariationHunter). We have also tested this algorithm on our own sequencing data, obtained from a tumour and a normal blood sample of a patient with chronic lymphocytic leukaemia, on which we have also validated the results by targeted re-sequencing of different kinds of predictions. This allowed us to determine confidence parameters that influence the reliability of breakpoint predictions.