ACNViewer: Comprehensive genome-wide visualization of absolute copy number and copy neutral variations.

Victor Renault,Sandrine Imbeaud,Fabien Pichon,Jean-François Deleuze,Jörg Tost,Eric Letouzé,Alexandre How-Kit,Jessica Zucman-Rossi,Shu-Fang Wang-Renault,Michael Baudis

doi:10.1371/journal.pone.0189334

Abstract

MotivationCopy number variations (CNV) include net gains or losses of part or whole chromosomal regions. They differ from copy neutral loss of heterozygosity (cn-LOH) events which do not induce any net change in the copy number and are often associated with uniparental disomy. These phenomena have long been reported to be associated with diseases and particularly in cancer. Losses/gains of genomic regions are often correlated with lower/higher gene expression. On the other hand, loss of heterozygosity (LOH) and cn-LOH are common events in cancer and may be associated with the loss of a functional tumor suppressor gene. Therefore, identifying recurrent CNV and cn-LOH events can be important as they may highlight common biological components and give insights into the development or mechanisms of a disease. However, no currently available tools allow a comprehensive whole-genome visualization of recurrent CNVs and cn-LOH in groups of samples providing absolute quantification of the aberrations leading to the loss of potentially important information.ResultsTo overcome these limitations, we developed aCNViewer (Absolute CNV Viewer), a visualization tool for absolute CNVs and cn-LOH across a group of samples. aCNViewer proposes three graphical representations: dendrograms, bi-dimensional heatmaps showing chromosomal regions sharing similar abnormality patterns, and quantitative stacked histograms facilitating the identification of recurrent absolute CNVs and cn-LOH. We illustrated aCNViewer using publically available hepatocellular carcinomas (HCCs) Affymetrix SNP Array data (Fig 1A). Regions 1q and 8q present a similar percentage of total gains but significantly different copy number gain categories (p-value of 0.0103 with a Fisher exact test), validated by another cohort of HCCs (p-value of 5.6e-7) (Fig 2B).Availability and implementationaCNViewer is implemented in python and R and is available with a GNU GPLv3 license on GitHub https://github.com/FJD-CEPH/aCNViewer and Docker https://hub.docker.com/r/fjdceph/acnviewer/.ContactaCNViewer@cephb.fr

Highlights

Human cancers can be characterized by different levels of genomic instability
We developed aCNViewer (Absolute Copy number variations (CNV) Viewer), a visualization tool for absolute CNVs and copy neutral loss of heterozygosity (cn-loss of heterozygosity (LOH)) across a group of samples. aCNViewer proposes three graphical representations: dendrograms, bi-dimensional heatmaps showing chromosomal regions sharing similar abnormality patterns, and quantitative stacked histograms facilitating the identification of recurrent absolute CNVs and cn-LOH
Results & discussion aCNViewer includes a whole pipeline for processing raw SNP and whole exome/genome paired bam data with ASCAT [22] and Sequenza [28] algorithms respectively which allow the identification of absolute copy number and copy neutral variations for each sample individually (Fig 3)

Summary

Introduction

Human cancers can be characterized by different levels of genomic instability. Whole or sections of chromosomes can be recurrently rearranged, lost, or amplified. We developed aCNViewer, a genome-wide visualization tool for the representation of absolute copy number and copy neutral variations (chromosome losses and gains, LOH and cnLOH) of groups of samples.

Results

Conclusion