GRAPES: A Versatile Tool for Analyzing Structural Variation From Whole‐Genome and Targeted DNA Sequencing Data

Abstract

The annotation of human genomic diversity is nowadays at an unprecedented rapid pace, but arguably still limited by the lack of computational approaches that integrate the analysis of multiple forms of genetic variation and are suitable for the analysis of both, whole‐genome sequencing and targeted sequencing strategies. For the analysis of structural variations (SVs), for example, which include >50 bp deletions, duplications, copy‐number variations (CNVs), inversions, mobile‐element insertions, and translocations, only a few computational tools integrate breakpoint signatures and read‐depth information in a single approach. And, none of them makes use, in the analysis of targeted sequencing, of breakpoint information to filter false‐positive CNVs when small adjacent introns are completely captured despite that targeted sequencing is still more popular than whole‐genome sequencing (e.g. exome‐sequencing and many common approaches of genetic‐testing are forms of targeted sequencing). Here, we present the development of GRAPES (Genomic Rearrangement Analyzer for Paired‐End Sequencing), which is as a versatile tool that allows the analysis of SVs on whole‐genome and targeted DNA sequencing data by integrating discordant‐pairs, split‐reads, read‐depth, and local assembly. In the case of targeted sequencing, GRAPES uses on‐ and off‐ target read depth coupled with breakpoint analysis to provide accurate predictions. We have validated GRAPES using the accurately annotated NA12878 genome, the ICR96 CNV exon validation series, and a panel of simulated datasets, and it achieves higher rates of sensitivity and specificity than previously developed approaches. We have tested GRAPES to annotate SVs from targeted sequencing data of patients affected with inherited cardiac disorders, uncovering complex SVs that were unresolved by using other algorithms. In sum, GRAPES not only is a comprehensive tool for the analysis of SVs from whole‐genome sequencing and targeted sequencing data, but also allows the annotation of SVs overlooked by current single‐approach approaches.Support or Funding InformationInternational Doctorate Grant in Biomedical Research and Public health of the Università Cattolica del Sacro CuoreThis abstract is from the Experimental Biology 2018 Meeting. There is no full text article associated with this abstract published in The FASEB Journal.